BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
X-LIC-LOCATION:Asia/Tokyo
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:JST
DTSTART:18871231T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260817T171604Z
LOCATION:Hall C\, C Block\, Level 4
DTSTART;TZID=Asia/Tokyo:20241203T090000
DTEND;TZID=Asia/Tokyo:20241203T120000
UID:siggraphasia_SIGGRAPH Asia 2024_sess223@linklings.com
SUMMARY:Technical Papers Fast-Forward and Awards
DESCRIPTION:A preview session of all Technical Papers will also be held on
  the first day of the event where author(s) of each paper get less than a 
 minute to wow the attendees with a brief overview of their work.\n\nExact 
 and Efficient Intersection Resolution for Mesh Arrangements\n\nWe propose 
 a novel method to exactly and efficiently resolve intersections and self-i
 ntersections in triangle meshes. Our method contains two key components. F
 irst, we present a new concept of geometric predicates, called indirect of
 fset predicates, to represent all intersection points through a new...\n\n
 \nJia-Peng Guo and Xiao-Ming Fu (University of Science and Technology of C
 hina)\n---------------------\nDesigning triangle meshes with controlled ro
 ughness\n\nMotivated by the emergence of rough surfaces in various areas o
 f design, we address the computational design of triangle meshes with cont
 rolled roughness. Our focus lies on small levels of roughness. There, roug
 hness or smoothness mainly arises through the local positioning of the mes
 h edges and fac...\n\n\nVictor Ceballos Inza (King Abdullah University of 
 Science and Technology (KAUST)); Panagiotis Fykouras (BMW Group); Florian 
 Rist (King Abdullah University of Science and Technology (KAUST), Technica
 l University of Vienna); Daniel Häseker and Majid Hojjat (BMW Group); Chri
 stian Müller (Technical University of Vienna); and Helmut Pottmann (Techni
 cal University of Vienna, King Abdullah University of Science and Technolo
 gy (KAUST))\n---------------------\nStripe Embedding: Efficient Maps with 
 Exact Numeric Computation\n\nWe consider the fundamental problem of inject
 ively mapping a surface mesh with disk topology onto a boundary constraine
 d convex domain. We start from the basic observation that mapping a strip 
 of triangles onto a rectangular shape always yields a valid embedding if t
 he vertices that bound the strip ...\n\n\nMarco Livesu (CNR IMATI)\n------
 ---------------\nBijective Volumetric Mapping via Star Decomposition\n\nA 
 method for the construction of bijective volumetric maps between 3D shapes
  is presented. Arbitrary shapes of ball-topology are supported, overcoming
  restrictions of previous methods to convex or star-shaped targets. In ess
 ence, the mapping problem is decomposed into a set of simpler mapping prob
 le...\n\n\nSteffen Hinderink, Hendrik Brückler, and Marcel Campen (Osnabrü
 ck University)\n---------------------\nPCO: Precision-Controllable Offset 
 Surfaces with Sharp Features\n\nSurface offsetting is a crucial operation 
 in digital geometry processing and computer-aided design, where an offset 
 is defined as an iso-value surface of the distance field. A challenge emer
 ges as even smooth surfaces can exhibit sharp features in their offsets du
 e to the non-differentiable characte...\n\n\nLei Wang (Shandong University
 , School of Computer Science and Technology); Xudong Wang and Pengfei Wang
  (Shandong University); Shuangmin Chen (Qingdao University of Science and 
 Technology); Shiqing Xin and Jiong Guo (Shandong University); Wenping Wang
  (Texas A&M University); and Changhe Tu (Shandong University)\n-----------
 ----------\nA Progressive Embedding Approach to Bijective Tetrahedral Maps
  driven by Cluster Mesh Topology\n\nWe present a novel algorithm to map ba
 ll-topology tetrahedral meshes onto star-shaped domains with guarantees re
 garding bijectivity. Our algorithm is based on the recently introduced ide
 a of Shrink-and-Expand, where images of interior vertices are initially cl
 ustered at one point (Shrink-), before b...\n\n\nValentin Zenon Nigolian (
 University of Bern), Marcel Campen (Osnabrück University), and David Bomme
 s (University of Bern)\n---------------------\nInfNeRF: Towards Infinite S
 cale NeRF Rendering with O(log n) Space Complexity\n\nThe conventional mes
 h-based Level of Detail (LoD) technique, exemplified by applications such 
 as Google Earth and many game engines, exhibits the capability to holistic
 ally represent a large scene even the Earth, and achieves rendering with a
  space complexity of O(log n).\nThis constrained data requi...\n\n\nJiabin
  Liang and Lanqing Zhang (Sea AI Lab), Zhuoran Zhao (National University o
 f Singapore), and Xiangyu Xu (Xi'an Jiaotong University)\n----------------
 -----\nTaming 3DGS: High-Quality Radiance Fields with Limited Resources\n\
 n3D Gaussian Splatting (3DGS) has transformed novel-view synthesis with it
 s fast, interpretable, and high-fidelity rendering. However, its resource 
 requirements limit its usability: Especially on weaker or constrained devi
 ces, training performance degrades quickly and often cannot complete due t
 o exc...\n\n\nSaswat Subhajyoti Mallick (Carnegie Mellon University); Rahu
 l Goel (International Institute of Information Technology, Hyderabad); Ber
 nhard Kerbl (Carnegie Mellon University); Markus Steinberger (Graz Univers
 ity of Technology); and Francisco Vicente Carrasco and Fernando De La Torr
 e (Carnegie Mellon University)\n---------------------\nRepresenting Long V
 olumetric Video with Temporal Gaussian Hierarchy\n\nThis paper aims to add
 ress the challenge of reconstructing long volumetric videos from multi-vie
 w RGB videos.\nRecent dynamic view synthesis methods leverage powerful 4D 
 representations, like feature grids or point cloud sequences, to achieve h
 igh-quality rendering results. However, they are typicall...\n\n\nZhen Xu 
 (State Key Laboratory of CAD&CG, Zhejiang University; Zhejiang University)
 ; Yinghao Xu (Stanford University); Zhiyuan Yu (Department of Mathematics,
  Hong Kong University of Science and Technology); Sida Peng and Jiaming Su
 n (Zhejiang University); and Hujun Bao and Xiaowei Zhou (State Key Laborat
 ory of CAD&CG, Zhejiang University)\n---------------------\nLetsGo: Large-
 Scale Garage Modeling and Rendering via LiDAR-Assisted Gaussian Primitives
 \n\nLarge garages are ubiquitous yet intricate scenes that present unique 
 challenges due to their monotonous colors, repetitive patterns, reflective
  surfaces, and transparent vehicle glass. Conventional Structure from Moti
 on (SfM) methods for camera pose estimation and 3D reconstruction often fa
 il in th...\n\n\nJiadi Cui (ShanghaiTech University, Stereye Inc.); Junmin
 g Cao (Shanghai Advanced Research Institute, Chinese Academy of Sciences; 
 University of Chinese Academy of Sciences); Fuqiang Zhao (ShanghaiTech Uni
 versity, NeuDim Inc.); Zhipeng He and Yifan Chen (ShanghaiTech University)
 ; Yuhui Zhong (DGene Inc.); Lan Xu and Yujiao Shi (ShanghaiTech University
 ); Yingliang Zhang (DGene Inc.); and Jingyi Yu (ShanghaiTech University)\n
 ---------------------\nMVImgNet2.0: A Larger-scale Dataset of Multi-view I
 mages\n\nMVImgNet is a large-scale dataset that contains multi-view images
  of ~220k real-world objects in 238 classes. As a counterpart of ImageNet,
  it introduces 3D visual signals via multi-view shooting, making a soft br
 idge between 2D and 3D vision. This paper constructs the MVImgNet2.0 datas
 et that expan...\n\n\nXiaoguang Han (SSE, The Chinese University of Hong K
 ong, Shenzhen; FNii, The Chinese University of Hong Kong, Shenzhen); Yushu
 ang Wu, Luyue Shi, Haolin Liu, Hongjie Liao, and Lingteng Qiu (FNii, The C
 hinese University of Hong Kong, Shenzhen; SSE, The Chinese University of H
 ong Kong, Shenzhen); Weihao Yuan, Xiaodong Gu, and Zilong Dong (Alibaba); 
 and Shuguang Cui (SSE, The Chinese University of Hong Kong, Shenzhen; FNii
 , The Chinese University of Hong Kong, Shenzhen)\n---------------------\nM
 oA: Mixture-of-Attention for Subject-Context Disentanglement in Personaliz
 ed Image Generation\n\nWe introduce a new architecture for personalization
  of text-to-image diffusion models, coined Mixture-of-Attention (MoA). Ins
 pired by the Mixture-of-Experts mechanism utilized in large language model
 s (LLMs), MoA distributes the generation workload between two attention pa
 thways: a personalized bran...\n\n\nKuan-Chieh Wang, Daniil Ostashev, Yuwe
 i Fang, Sergey Tulyakov, and Kfir Aberman (Snap Inc.)\n-------------------
 --\nReVersion: Diffusion-Based Relation Inversion from Images\n\nDiffusion
  models gain increasing popularity for their generative capabilities. Rece
 ntly, there have been surging needs to generate customized images by inver
 ting diffusion models from exemplar images, and existing inversion methods
  mainly focus on capturing object appearances (i.e., the "look"). How...\n
 \n\nZiqi Huang, Tianxing Wu, Yuming Jiang, Kelvin C.K. Chan, and Ziwei Liu
  (S-Lab for Advanced Intelligence, Nanyang Technological University Singap
 ore)\n---------------------\nPALP: Prompt Aligned Personalization of Text-
 to-Image Models\n\nContent creators often aim to create personalized image
 s using personal subjects that go beyond the capabilities of conventional 
 text-to-image models. Additionally, they may want the resulting image to e
 ncompass a specific location, style, ambiance, and more. Existing personal
 ization methods may com...\n\n\nMoab Arar (Tel Aviv University), Andrey Vo
 ynov and Amir Hertz (Google Research), Omri Avrahami (Hebrew University of
  Jerusalem), Shlomi Fruchter and Yael Pritch (Google Research), Daniel Coh
 en-Or (Tel Aviv University), and Ariel Shamir (Reichman University)\n-----
 ----------------\nCustomizing Text-to-Image Models with a Single Image Pai
 r\n\nArt reinterpretation is the practice of creating a variation of a ref
 erence work, making a paired artwork that exhibits a distinct artistic sty
 le. We ask if such an image pair can be used to customize a generative mod
 el to capture the demonstrated stylistic difference. We propose Pair Custo
 mization,...\n\n\nMaxwell Jones, Sheng-Yu Wang, and Nupur Kumari (Carnegie
  Mellon University); David Bau (Northeastern University); and Jun-Yan Zhu 
 (Carnegie Mellon University)\n---------------------\nCustomizing Text-to-I
 mage Diffusion with Object Viewpoint Control\n\nModel customization introd
 uces new concepts to existing text-to-image models, enabling the generatio
 n of these new concepts/objects in novel contexts.\nHowever, such methods 
 lack accurate camera view control with respect to the new object, and user
 s must resort to prompt engineering (e.g., adding "to...\n\n\nNupur Kumari
  and Grace Su (Carnegie Mellon Uniersity); Richard Zhang, Taesung Park, an
 d Eli Shechtman (Adobe Research); and Jun-Yan Zhu (Carnegie Mellon Uniersi
 ty)\n---------------------\nIdentity-Preserving Face Swapping via Dual Sur
 rogate Generative Models\n\nIn this study, we revisit the fundamental sett
 ing of face-swapping models and reveal that only using implicit supervisio
 n for training leads to the difficulty of advanced methods to preserve the
  source identity. We propose a novel reverse pseudo-input generation appro
 ach to offer supplemental data f...\n\n\nZiyao Huang and Fan Tang (Institu
 te of Computing Technology, Chinese Academy of Sciences); Yong Zhang (Tenc
 ent); Juan Cao, Chengyu Li, Sheng Tang, and Jintao Li (Institute of Comput
 ing Technology, Chinese Academy of Sciences); and Tong-Yee Lee (National C
 heng Kung University)\n---------------------\nHFH-Font: Few-shot Chinese F
 ont Synthesis with Higher Quality, Faster Speed, and Higher Resolution\n\n
 The challenge of automatically synthesizing high-quality vector fonts, par
 ticularly for writing systems (e.g., Chinese) consisting of huge amounts o
 f complex glyphs, remains unsolved. Existing font synthesis techniques fal
 l into two categories: 1) methods that directly generate vector glyphs, an
 d 2)...\n\n\nHua Li (Wangxuan Institute of Computer Technology, Peking Uni
 versity) and Zhouhui Lian (Wangxuan Institute of Computer Technology, Peki
 ng University; State Key Laboratory of General Artificial Intelligence, Pe
 king University)\n---------------------\nSD-𝜋XL: Generating Low-Resolution
  Quantized Imagery via Score Distillation\n\nLow-resolution quantized imag
 ery, such as pixel art, is seeing a revival in modern applications ranging
  from video game graphics to digital design and fabrication, where creativ
 ity is often bound by a limited palette of elemental units. Despite their 
 growing popularity, the automated generation of q...\n\n\nAlexandre Binnin
 ger and Olga Sorkine-Hornung (ETH Zürich)\n---------------------\nProcessP
 ainter: Learning to draw from sequence data\n\nThe painting process of art
 ists is inherently stepwise and varies significantly among different paint
 ers and styles. Generating detailed, step-by-step painting processes is es
 sential for art education and research, yet remains largely underexplored.
  Traditional stroke-based rendering methods break d...\n\n\nYiren Song (Na
 tional University of Singapore, Show Lab); Shijie Huang, Chen Yao, and Hai
  Ci (National University of Singapore); Xiaojun Ye (Zhejiang University); 
 Jiaming Liu (Tiamat); Yuxuan Zhang (Shanghai Jiao Tong University); and Mi
 ke Zheng Shou (National University of Singapore)\n---------------------\nI
 nverse Painting: Reconstructing The Painting Process\n\nGiven an input pai
 nting, we reconstruct a time-lapse video of how it may be painted.  We for
 mulate this as an autoregressive image generation problem, in which an ini
 tially blank "canvas'' is iteratively updated. The model learns from real 
 artists by training on many painting videos.\nOur approach in...\n\n\nBowe
 i Chen, Yifan Wang, Brian Curless, Ira Kemelmacher-Shlizerman, and Steven 
 M. Seitz (University of Washington)\n---------------------\nLVCD: Referenc
 e-based Lineart Video Colorization with Diffusion Models\n\nWe propose the
  first video diffusion framework for reference-based lineart video coloriz
 ation. Unlike previous works that rely solely on image generative models t
 o colorize lineart frame by frame, our approach leverages a large-scale pr
 etrained video diffusion model to generate colorized animation v...\n\n\nZ
 hitong Huang (City University of Hong Kong); Mohan Zhang (Wechat, Tencent 
 Inc.); and Jing Liao (City University of Hong Kong)\n---------------------
 \nColorful Diffuse Intrinsic Image Decomposition in the Wild\n\nIntrinsic 
 image decomposition aims to separate the surface reflectance and the effec
 ts from the illumination given a single photograph. Due to the complexity 
 of the problem, most prior works assume a single-color illumination and a 
 Lambertian world, which limits their use in illumination-aware image...\n\
 n\nChris Careaga and Yağız Aksoy (Simon Fraser University)\n--------------
 -------\nDifferential Walk on Spheres\n\nWe introduce a Monte Carlo method
  for computing derivatives of the solution to a partial differential equat
 ion (PDE) with respect to problem parameters (such as domain geometry or b
 oundary conditions). Derivatives can be evaluated at arbitrary points, wit
 hout performing a global solve or constructin...\n\n\nBailey Miller (Carne
 gie Mellon Uniersity), Rohan Sawhney (NVIDIA), and Keenan Crane and Ioanni
 s Gkioulekas (Carnegie Mellon Uniersity)\n---------------------\nProjected
  Walk on Spheres: A Monte Carlo Closest Point Method for Surface PDEs\n\nW
 e present projected walk on spheres (PWoS), a novel pointwise and discreti
 zation-free Monte Carlo solver for surface PDEs with Dirichlet boundaries,
  as a generalization of the walk on spheres method (WoS)  [Muller 1956; Sa
 whney and Crane 2020]. We adapt the recursive relationship of WoS designed
  fo...\n\n\nRyusuke Sugimoto, Nathan King, Toshiya Hachisuka, and Christop
 her Batty (University of Waterloo)\n---------------------\nSolving Inverse
  PDE Problems using Grid-Free Monte Carlo Estimators\n\nPartial differenti
 al equations can model diverse physical phenomena including heat diffusion
 , incompressible flows, and electrostatic potentials. Given a description 
 of an object's boundary and interior, traditional methods solve such PDEs 
 by densely meshing the interior and then solving a large and...\n\n\nEkrem
  Fatih Yilmazer (EPFL), Delio Vicini (Google Inc.), and Wenzel Jakob (EPFL
 )\n---------------------\nDiffCSG: Differentiable CSG via Rasterization\n\
 nDifferentiable rendering is a key ingredient for inverse rendering and ma
 chine learning, as it allows to optimize scene parameters (shape, material
 s, lighting) to best fit target images. Differentiable rendering requires 
 that each scene parameter relates to pixel values through differentiable o
 perat...\n\n\nHaocheng Yuan (University of Edinburgh); Adrien Bousseau (In
 ria Sophia-Antipolis, Université Côte d’Azur); Hao Pan (Microsoft Research
  Asia); Quancheng Zhang (Nanjing University); Niloy J. Mitra (University C
 ollege London (UCL), Adobe Research); and Changjian Li (University of Edin
 burgh)\n---------------------\nText-guided Controllable Mesh Refinement fo
 r Interactive 3D Modeling\n\nWe propose a novel technique for adding geome
 tric details to an input coarse 3D mesh guided by a text prompt. Our metho
 d is composed of three stages. First, we generate a single-view RGB image 
 conditioned on the input coarse geometry and the input text prompt. This s
 ingle-view image generation step ...\n\n\nYun-Chun Chen and Selena Ling (U
 niversity of Toronto); Zhiqin Chen, Vladimir G. Kim, and Matheus Gadelha (
 Adobe Research); and Alec Jacobson (University of Toronto)\n--------------
 -------\nI❤️MESH: A DSL for Mesh Processing\n\nMesh processing algorithms 
 are often communicated via concise mathematical notation (e.g., summation 
 over mesh neighborhoods). However, conversion of notation into working cod
 e remains a time-consuming and error-prone process, which requires arcane 
 knowledge of low-level data structures and librarie...\n\n\nYong Li (South
  China University of Technology, George Mason University); Shoaib Kamil (A
 dobe Research); Keenan Crane (Carnegie Mellon University); Alec Jacobson (
 University of Toronto, Adobe Research); and Yotam Gingold (George Mason Un
 iversity)\n---------------------\nDifFRelight: Diffusion-Based Facial Perf
 ormance Relighting\n\nWe present a novel framework for free-viewpoint faci
 al performance relighting using diffusion-based image-to-image translation
 . Leveraging a subject-specific dataset containing diverse facial expressi
 ons captured under various lighting conditions, including flat-lit and one
 -light-at-a-time (OLAT) sc...\n\n\nMingming He (Netflix Eyeline Studios); 
 Pascal Clausen (Netflix Eyeline Studios, Osylum); and Ahmet Levent Taşel, 
 Li Ma, Oliver Pilarski, Wenqi Xian, Laszlo Rikker, Xueming Yu, Ryan Burger
 t, Ning Yu, and Paul Debevec (Netflix Eyeline Studios)\n------------------
 ---\nGS^3: Efficient Relighting with Triple Gaussian Splatting\n\nWe prese
 nt a spatial and angular Gaussian based representation and a triple splatt
 ing process, for real-time, high-quality novel lighting-and-view synthesis
  from multi-view point-lit input images. To describe complex appearance, w
 e employ a Lambertian plus a mixture of angular Gaussians as an effect...\
 n\n\nZoubin Bi, Yixin Zeng, Chong Zeng, Fan Pei, Xiang Feng, Kun Zhou, and
  Hongzhi Wu (State Key Laboratory of CAD&CG, Zhejiang University)\n-------
 --------------\nOLAT Gaussians for Generic Relightable Appearance Acquisit
 ion\n\nOne-light-at-a-time (OLAT) images sample a broader range of object 
 appearance changes than images captured under constant lighting and are su
 perior as input to object relighting. Although existing methods have produ
 ced reasonable relighting quality using OLAT images, they utilize surface-
 like repres...\n\n\nZhiyi Kuang (State Key Laboratory of CAD&CG, Zhejiang 
 University); Yanchao Yang and Siyan Dong (University of Hong Kong); Jiayue
  Ma (State Key Laboratory of CAD&CG, Zhejiang University); Hongbo Fu (Hong
  Kong University of Science and Technology); and Youyi Zheng (State Key La
 boratory of CAD&CG, Zhejiang University)\n---------------------\nReflectio
 n-Aware Neural Radiance Fields\n\nNeural Radiance Fields (NeRF) have demon
 strated exceptional capabilities in reconstructing complex scenes with hig
 h fidelity. However, NeRF's view dependency can only handle low-frequency 
 reflections. It falls short when handling complex planar reflections, ofte
 n interpreting them as erroneous scene...\n\n\nChen Gao, Yipeng Wang, and 
 Changil Kim (Meta); Jia-Bin Huang (University of Maryland, College Park); 
 and Johannes Kopf (Meta)\n---------------------\nNeRF-Casting: Improved Vi
 ew-Dependent Appearance with Consistent Reflections\n\nNeural Radiance Fie
 lds (NeRFs) typically struggle to reconstruct and render highly specular o
 bjects, whose appearance varies quickly with changes in viewpoint. Recent 
 works have improved NeRF's ability to render detailed specular appearance 
 of distant environment illumination, but are unable to synt...\n\n\nDor Ve
 rbin, Pratul P. Srinivasan, Peter Hedman, and Ben Mildenhall (Google Resea
 rch); Benjamin Attal (Carnegie Mellon University); and Richard Szeliski an
 d Jonathan T. Barron (Google Research)\n---------------------\nLocal Gauss
 ian Density Mixtures for Unstructured Lumigraph Rendering\n\nTo improve no
 vel-view synthesis of curved surface reflections and refractions, we revis
 it local geometry-guided ray interpolation techniques with modern differen
 tiable rendering and optimization.\nIn contrast to depth or mesh geometrie
 s, our approach uses a local or per-view density represented as Ga...\n\n\
 nXiuchao Wu (State Key Laboratory of CAD&CG, Zhejiang University); Jiamin 
 Xu (Hangzhou Dianzi Univeristy); Chi Wang (State Key Laboratory of CAD&CG,
  Zhejiang University); Yifan Peng (University of Hong Kong); Qixing Huang 
 (University of Texas at Austin); James Tompkin (Brown University); and Wei
 wei Xu (State Key Laboratory of CAD&CG, Zhejiang University)\n------------
 ---------\nCOMFI: A Calibrated Observer Metameric Failure Index for Color 
 Critical Tasks\n\nIn imaging, various tasks require a comparison of images
  between displays with different emission spectra. Even if the displays ar
 e calibrated and predicted to match using traditional colorimetry, individ
 ual differences in color perception of observers can cause colors on two d
 isplays to appear mism...\n\n\nRobert Wanat (LG Electronics USA), Michael 
 D. Smith (Wavelet Consulting LLC), Junwoo Jang (LG Display), and Sally Hat
 tori (Walt Disney Studios)\n---------------------\nLarge Étendue 3D Hologr
 aphic Display with Content-adaptive Dynamic Fourier Modulation\n\nEmerging
  holographic display technology offers unique capabilities for next-genera
 tion virtual reality systems. Current holographic near-eye displays, howev
 er, only support a small etendue, which results in a direct tradeoff betwe
 en achievable field of view and eyebox size. Etendue expansion has rec...\
 n\n\nBrian Chao, Manu Gopakumar, and Suyeon Choi (Stanford University); Jo
 nghyun Kim (NVIDIA); Liang Shi (Massachusetts Institute of Technology (MIT
 )); and Gordon Wetzstein (Stanford University)\n---------------------\nela
 TCSF: A Temporal Contrast Sensitivity Function for Flicker Detection and M
 odeling Variable Refresh Rate Flicker\n\nThe perception of flicker has bee
 n a prominent concern in illumination and electronic display fields for ov
 er a century. Traditional approaches often rely on Critical Flicker Freque
 ncy (CFF), primarily suited for high-contrast (full-on, full-off) flicker.
  To tackle varying contrast flicker, the Inte...\n\n\nYancheng Cai (Univer
 sity of Cambridge), Ali Bozorgian (Norwegian University of Science and Tec
 hnology), Maliha Ashraf (University of Cambridge), Robert Wanat (LG Electr
 onics North America), and Rafal Mantiuk (University of Cambridge)\n-------
 --------------\nPerspective-Aligned AR Mirror with Under-Display Camera\n\
 nAugmented reality (AR) mirrors are novel displays that have great potenti
 al for commercial applications such as virtual apparel try-on. Typically t
 he camera is placed beside the display, leading to distorted perspectives 
 during user interaction. In this paper, we present a novel approach to add
 ress ...\n\n\nJian Wang, Sizhuo Ma, Karl Bayer, Yi Zhang, Peihao Wang, and
  Bing Zhou (Snap Inc.); Shree Nayar (Columbia University); and Gurunandan 
 Krishnan (Snap Inc.)\n---------------------\nAR-DAVID: Augmented Reality D
 isplay Artifact Video Dataset\n\nThe perception of visual content in optic
 al-see-through augmented reality (AR) devices is affected by the light com
 ing from the environment. This additional light interacts with the content
  in a non-trivial manner because of the illusion of transparency, differen
 t focal depths, and motion parallax. ...\n\n\nAlexandre Chapiro (Reality L
 abs, Meta); Dongyeon Kim (University of Cambridge); Yuta Asano (Reality La
 bs, Meta); and Rafał K. Mantiuk (University of Cambridge)\n---------------
 ------\nV^3: Viewing Volumetric Videos on Mobiles via Streamable 2D Dynami
 c Gaussians\n\nExperiencing high-fidelity volumetric video as seamlessly a
 s 2D videos is a long-held dream. However, current dynamic 3DGS methods, d
 espite their high rendering quality, face challenges in streaming on mobil
 e devices due to computational and bandwidth constraints. In this paper, w
 e introduce V3 (Vie...\n\n\nPenghao Wang, Zhirui Zhang, Liao Wang, Kaixin 
 Yao, and Siyuan Xie (ShanghaiTech University, NeuDim Inc.); Jingyi Yu (Sha
 nghaiTech University); Minye Wu (KU Leuven); and Lan Xu (ShanghaiTech Univ
 ersity)\n---------------------\nApproximation by Meshes with Spherical Fac
 es\n\nMeshes with spherical faces and circular edges are an attractive alt
 ernative to polyhedral meshes for applications in architecture and design.
  Approximation of a given surface by such a mesh needs to consider the vis
 ual appearance, approximation quality, the position and orientation of cir
 cular inte...\n\n\nAnthony S. Cisneros Ramos and Alisher Aikyn (King Abdul
 lah University of Science and Technology (KAUST)); Martin Kilian (TU Wien)
 ; Helmut Pottmann (TU Wien, King Abdullah University of Science and Techno
 logy (KAUST)); and Christian Müller (TU Wien)\n---------------------\nComp
 utational Biomimetics of Winged Seeds\n\nWe develop a computational pipeli
 ne to facilitate the biomimetic design of winged seeds. Our approach lever
 ages 3D scans of natural winged seeds to construct a bio-inspired design s
 pace by interpolating them with geodesic coordinates in the 3D diffeomorph
 ism group. We formulate aerodynamic design ta...\n\n\nQiqin Le (Shanghai Q
 i Zhi Institute); Jiamu Bu (Tsinghua University); Yanke Qu (Peking Univers
 ity); Bo Zhu (Georgia Institute of Technology); and Tao Du (Tsinghua Unive
 rsity, Shanghai Qi Zhi Institute)\n---------------------\nOptimized shock-
 protecting microstructures\n\nMechanical shock is a common occurrence in v
 arious settings, there are two different scenarios for shock protection: c
 atastrophic protection (e.g. car collisions and falls) and routine protect
 ion (e.g. shoe soles and mattresses). The former protects against one-time
  events, the latter against period...\n\n\nZizhou Huang, Daniele Panozzo, 
 and Denis Zorin (New York University)\n---------------------\nAll you need
  is rotation: Construction of developable strips\n\nWe present a novel app
 roach to generate developable strips along a space curve. The key idea of 
 the new method is to use the rotation angle between the Frenet frame of th
 e input space curve, and its Darboux frame of the curve on the resulting d
 evelopable strip as a free design parameter, thereby rev...\n\n\nTakashi M
 aekawa (Waseda University) and Felix Scholz (Johannes Kepler University Li
 nz)\n---------------------\nAlignable Lamella Gridshells\n\nAlignable lame
 lla gridshells are 3D grid structures capable of collapsing into a planar 
 strip. \nThis feature significantly simplifies on-site assembly and also e
 nsures compactness for efficient transport and storage.\nHowever, designin
 g these structures to achieve specific shapes still remains a chal...\n\n\
 nDavide Pellis (ISTI-CNR)\n---------------------\nLayout-Aware Single-Imag
 e Document Flattening\n\nOur paper presents a novel framework for layout-a
 ware single-image document flattening. It combines layout-aware segmentati
 on, local and global UV map prediction, and an merging algorithm. Addition
 ally, we introduce a new synthetic dataset containing meshes generated by 
 a physics-based simulation, e...\n\n\nPu Li, Jianwei Guo, Weize Quan, and 
 Dongming Yan (MAIS, Institute of Automation, CAS School of Artificial Inte
 lligence, UCAS)\n---------------------\nMARS: Multi-sample Allocation thro
 ugh Russian roulette and Splitting\n\nMultiple importance sampling (MIS) i
 s an indispensable tool in rendering that constructs robust sampling strat
 egies by combining the respective strengths of individual distributions. I
 ts efficiency can be greatly improved by carefully selecting the number of
  samples drawn from each distribution, but...\n\n\nJoshua Meyer, Alexander
  Rath, and Ömercan Yazici (Saarland Informatics Campus) and Philipp Slusal
 lek (German Research Center for Artificial Intelligence, Saarland Informat
 ics Campus)\n---------------------\nVolume Scattering Probability Guiding\
 n\nSimulating the light transport of volumetric effects poses significant 
 challenges and costs, especially in the presence of heterogeneous volumes.
  Generating stochastic paths for volume rendering involves multiple decisi
 ons, and previous works mainly focused on directional and distance samplin
 g, wher...\n\n\nKehan Xu (ETH Zürich); Sebastian Herholz (Intel Corporatio
 n); Marco Manzi and Marios Papas (DisneyResearch|Studios); and Markus Gros
 s (DisneyResearch|Studios, ETH Zürich)\n---------------------\nEfficient N
 eural Path Guiding with 4D Modeling\n\nPrevious local guiding methods used
  3D data structures to model spatial radiance variations but struggled wit
 h additional dimensions in the path integral, such as temporal changes in 
 dynamic scenes. Extending these structures to higher dimensions also prove
 s inefficient due to the curse of dimension...\n\n\nHonghao Dong, Rui Su, 
 Guoping Wang, and Sheng Li (Peking University)\n---------------------\nNeu
 Smoke: Efficient Smoke Reconstruction and View Synthesis with Neural Trans
 portation Fields\n\nNovel view synthesis of smoke scenes presents a challe
 nging problem. Previous neural approaches have suffered from inadequate qu
 ality and inefficient training. We introduce NeuSmoke, an efficient framew
 ork for dynamic smoke reconstruction using neural transportation fields, e
 nabling high-quality den...\n\n\nJiaxiong Qiu (TMCC, College of Computer S
 cience, Nankai University; Horizon Robotics); Ruihong Cen (TMCC, College o
 f Computer Science, Nankai University); Zhong Li (Apple); Han Yan (Nankai 
 TMCC, College of Computer Science, Nankai University); and Ming-Ming Cheng
  and Bo Ren (TMCC, College of Computer Science, Nankai University)\n------
 ---------------\nDynamic Neural Radiosity with Multi-grid Decomposition\n\
 nPrior approaches to the neural rendering of global illumination typically
  rely on complex network architectures and training strategies to model th
 e global effects. This often leads to impractically high overheads for bot
 h training and inference. The neural radiosity technique marks a significa
 nt ad...\n\n\nRui Su, Honghao Dong, Jierui Ren, Haojie Jin, Yisong Chen, G
 uoping Wang, and Sheng Li (Peking University)\n---------------------\nNeur
 al Global Illumination via Superposed Deformable Feature Fields\n\nInterac
 tive rendering of dynamic scenes with complex global illumination has been
  a long-standing problem in computer graphics.\nRecent advances in neural 
 rendering demonstrate new promising possibilities.\nHowever, while existin
 g methods have achieved impressive results, complex rendering effects (e..
 ..\n\n\nChuankun Zheng, Yuchi Huo, Hongxiang Huang, and Hongtao Sheng (Sta
 te Key Laboratory of CAD&CG, Zhejiang University); Junrong Huang (City Uni
 versity of Hong Kong); Rui Tang and Hao Zhu (Manycore Inc.); and Rui Wang 
 and Hujun Bao (State Key Laboratory of CAD&CG, Zhejiang University)\n-----
 ----------------\nQuark: Real-time, High-resolution, and General Neural Vi
 ew Synthesis\n\nWe present a novel neural algorithm for performing high-qu
 ality, high-resolution, real-time novel view synthesis. From a sparse set 
 of input RGB images or videos streams, our network both reconstructs the 3
 D scene and renders novel views at 1080p resolution at 30fps on an NVIDIA 
 A100. Our feed-forwa...\n\n\nJohn Flynn, Michael Broxton, Lukas Murmann, L
 ucy Chai, Matthew DuVall, Clément Godard, Kathryn Heal, Srinivas Kaza, Ste
 phen Lombardi, Xuan Luo, Supreeth Achar, Kira Prabhu, Tiancheng Sun, Lynn 
 Tsai, and Ryan Overbeck (Google)\n---------------------\nPano2Room: Novel 
 View Synthesis from a Single Indoor Panorama\n\nRecent single-view 3D AIGC
  methods have made significant advancements by leveraging knowledge distil
 led from extensive 3D object datasets. However, challenges persist in the 
 synthesis of 3D scenes from a single view, primarily due to the complexity
  of real-world environments and the limited availabi...\n\n\nGuo Pu, Yimin
 g Zhao, and Zhouhui Lian (Wangxuan Institute of Computer Technology, Pekin
 g University)\n---------------------\nCafca: High-quality Novel View Synth
 esis of Expressive Faces from Casual Few-shot Captures\n\nVolumetric model
 ing and neural radiance field representations have revolutionized 3D face 
 capture and photorealistic novel view synthesis. However, these methods of
 ten require hundreds of multi-view input images and are thus inapplicable 
 to cases with less than a handful of inputs.\nWe present a nove...\n\n\nMa
 rcel C. Buehler and Gengyan Li (ETH Zürich, Google VR); Erroll Wood, Leonh
 ard Helminger, Xu Chen, Tanmay Shah, Daoye Wang, Stephan Garbin, and Sergi
 o Orts Escolano (Google VR); Otmar Hilliges (ETH Zürich); and Dmitry Lagun
 , Jérémy Riviere, Paulo Gotardo, Thabo Beeler, Abhimitra Meka, and Kripasi
 ndhu Sarkar (Google VR)\n---------------------\nDynamic Gaussian Marbles f
 or Novel View Synthesis of Casual Monocular Videos\n\nGaussian splatting h
 as become a popular representation for novel-view synthesis, exhibiting cl
 ear strengths in efficiency, photometric quality, and compositional edibil
 ity. Following its success, many works have extended Gaussians to 4D, show
 ing that dynamic Gaussians maintain these benefits while a...\n\n\nColton 
 Stearns, Adam Harley, and Mikaela Uy (Stanford University); Florian Dubost
  and Federico Tombari (Google Research); and Gordon Wetzstein and Leonidas
  Guibas (Stanford University)\n---------------------\nNeural Light Spheres
  for Implicit Image Stitching and View Synthesis\n\nChallenging to capture
 , and challenging to display on a cellphone screen, the panorama paradoxic
 ally remains both a staple and underused feature of modern mobile camera a
 pplications. In this work we address both of these challenges with a spher
 ical neural light field model for implicit panoramic ima...\n\n\nIlya Chug
 unov and Amogh Joshi (Princeton University), Kiran Murthy and Francois Ble
 ibel (Google Inc.), and Felix Heide (Princeton University)\n--------------
 -------\nReN Human: Learning Relightable Neural Implicit Surfaces for Anim
 atable Human Rendering\n\nThis work proposes ReN Human, a framework that u
 tilizes sparse or even monocular input videos to reconstruct a 3D human mo
 del represented as a deformable implicit neural surface. It decomposes geo
 metry and material, resulting in a relightable, animatable human model tha
 t can be rendered with novel v...\n\n\nRengan Xie (State Key Laboratory of
  CAD&CG, Zhejiang University); Kai Huang (Institute of Computing Technolog
 y, Chinese Academy of Sciences; Zhejiang Lab); In-Young Cho (KRAFTON); Sen
  Yang (Zhejiang Lab); Wei Chen, Hujun Bao, and Wenting Zheng (State Key La
 boratory of CAD&CG, Zhejiang University); Rong Li (Zhejiang University); a
 nd Yuchi Huo (State Key Laboratory of CAD&CG, Zhejiang University; Zhejian
 g Lab)\n---------------------\nStochastic Normal Orientation for Point Clo
 uds\n\nWe propose a simple yet effective method to orient normals for poin
 t clouds. Central to our approach is a novel optimization objective functi
 on defined from global and local perspectives. Globally, we introduce a si
 gned uncertainty function that distinguishes the inside and outside of the
  underlying...\n\n\nGuojin Huang, Qing Fang, Zheng Zhang, Ligang Liu, and 
 Xiao-Ming Fu (University of Science and Technology of China)\n------------
 ---------\nFast and Globally Consistent Normal Orientation based on the Wi
 nding Number Normal Consistency\n\nEstimating consistently oriented normal
 s for point clouds enables a number of important applications in computer 
 graphics such as surface reconstruction. While local normal estimation is 
 possible with simple techniques like principal component analysis (PCA), o
 rienting these normals to be globally c...\n\n\nSiyou Lin (Department of A
 utomation, Tsinghua University); Zuoqiang Shi (Yau Mathematical Sciences C
 enter, Tsinghua University; Yanqi Lake Beijing Institute of Mathematical S
 ciences and Applications); and Yebin Liu (Department of Automation, Tsingh
 ua University)\n---------------------\nSurface Reconstruction Using Rotati
 on Systems\n\nInspired by the seminal result that a graph and an associate
 d rotation system uniquely determine the topology of a closed manifold, we
  propose a combinatorial method for reconstruction of surfaces from points
 . Our method constructs a spanning tree and a rotation system. Since the t
 ree is trivially a...\n\n\nRuiqi Cui, Emil Toftegaard Gæde, and Eva Rotenb
 erg (Technical University of Denmark); Leif Kobbelt (Visual Computing Inst
 itute, RWTH Aachen University); and J. Andreas Bærentzen (Technical Univer
 sity of Denmark)\n---------------------\nPVP-Recon: Progressive View Plann
 ing via Warping Consistency for Sparse-View Surface Reconstruction\n\nNeur
 al implicit representations have revolutionized dense multi-view surface r
 econstruction, yet their performance significantly diminishes with sparse 
 input views. A few pioneering works have sought to tackle the challenge of
  sparse-view reconstruction by leveraging additional geometric priors or m
 ...\n\n\nSheng Ye, Yuze He, Matthieu Lin, Jenny Sheng, and Ruoyu Fan (Tsin
 ghua University); Yiheng Han (Beijing University of Technology); Yubin Hu 
 (Tsinghua University); Ran Yi (Shanghai Jiao Tong University); Yu-Hui Wen 
 (Beijing Jiaotong University); Yong-Jin Liu (Tsinghua University); and Wen
 ping Wang (Texas A&M University)\n---------------------\n3D Reconstruction
  with Fast Dipole Sums\n\nWe introduce a method for high-quality 3D recons
 truction from multi-view images. Our method uses a new point-based represe
 ntation, the regularized dipole sum, which generalizes the winding number 
 to allow for interpolation of per-point attributes in point clouds with no
 isy or outlier points. Using r...\n\n\nHanyu Chen, Bailey Miller, and Ioan
 nis Gkioulekas (Carnegie Mellon University)\n---------------------\nArchit
 ectural Co-LOD Generation\n\nManaging the level-of-detail (LOD) in archite
 ctural models is crucial yet challenging, particularly for effective repre
 sentation and visualization of buildings. Traditional approaches often fai
 l to deliver controllable detail alongside semantic consistency, especiall
 y when dealing with noisy and inc...\n\n\nRunze Zhang, Shanshan Pan, and C
 henlei Lv (Shenzhen University (SZU)); Minglun Gong (University of Guelph)
 ; and Hui Huang (Shenzhen University (SZU))\n---------------------\nLLM-en
 hanced Scene Graph Learning for Household Rearrangement\n\nThe household r
 earrangement task involves spotting misplaced objects in a scene and accom
 modate  them with proper places. It depends both on common-sense knowledge
  on the objective side and human user preference on the subjective side. I
 n achieving such task, we propose to mine object functionality ...\n\n\nWe
 nhao Li, Zhiyuan Yu, Qijin She, Zhinan Yu, Yuqing Lan, and Chenyang Zhu (N
 ational University of Defense Technology (NUDT)); Ruizhen Hu (Shenzhen Uni
 versity (SZU)); and Kai Xu (National University of Defense Technology (NUD
 T))\n---------------------\nSGEdit: Bridging LLM with Text2Image Generativ
 e Model for Scene Graph-based Image Editing\n\nScene graphs offer a struct
 ured, hierarchical representation of images, with nodes and edges symboliz
 ing objects and the relationships among them. It can serve as a natural in
 terface for image editing, dramatically improving precision and flexibilit
 y. Leveraging this benefit, we introduce a new fram...\n\n\nZhiyuan Zhang 
 (City University of Hong Kong), DongDong Chen (Microsoft GenAI), and Jing 
 Liao (City University of Hong Kong)\n---------------------\nCPoser: An Opt
 imization-after-Parsing Approach for Text-to-Pose Generation Using Large L
 anguage Models.\n\nText-to-pose generation is challenging due to the compl
 exity of natural language and human posture semantics. Utilizing large lan
 guage models (LLMs) for text-to-pose generation is appealing due to their 
 strong capabilities in text understanding and reasoning. However, as LLMs 
 are designed for genera...\n\n\nYumeng Li, Bohong Chen, Zhong Ren, and Yao
 -Xiang Ding (Zhejiang University); Libin Liu (Peking University); and Tian
 jia Shao and Kun Zhou (Zhejiang University)\n---------------------\nParSEL
 : Parameterized Shape Editing with Language\n\nThe ability to edit 3D asse
 ts from natural language presents a compelling paradigm to aid in the demo
 cratization of 3D content creation. However, while natural language is oft
 en effective at communicating general intent, it is poorly suited for spec
 ifying exact manipulation. To address this gap, we ...\n\n\nAditya Ganesha
 n, Ryan Huang, Xianghao Xu, R. Kenny Jones, and Daniel Ritchie (Brown Univ
 ersity)\n---------------------\nAutonomous Character-Scene Interaction Syn
 thesis from Text Instruction\n\nSynthesizing human motions in 3D environme
 nts, particularly those with complex activities such as locomotion, hand-r
 eaching, and human-object interaction, presents substantial demands for us
 er-defined waypoints and stage transitions. These requirements pose challe
 nges for current models, leading to ...\n\n\nNan Jiang (Peking University,
  Beijing Institute for General Artificial Intelligence); Zimo He (Peking U
 niversity); Zi Wang (Beijing University of Posts and Telecommunications); 
 Hongjie Li (Peking University); Yixin Chen and Siyuan Huang (Beijing Insti
 tute for General Artificial Intelligence); and Yixin Zhu (Peking Universit
 y)\n---------------------\nAnim-Director: A Large Multimodal Model Powered
  Agent for Controllable Animation Video Generation\n\nTraditional animatio
 n generation methods depend on training generative models with human-label
 led data, entailing a sophisticated multi-stage pipeline that demands subs
 tantial human effort and incurs high training costs. Due to limited prompt
 ing plans, these methods typically produce brief, informat...\n\n\nYunxin 
 Li, Haoyuan Shi, and Baotian Hu (Harbin Institute of Technology); Longyue 
 Wang (Alibaba Group); Jiashun Zhu and Jinyi Xu (Jilin University); Zhen Zh
 ao (Tencent AILab); and Min Zhang (Harbin Institute of Technology)\n------
 ---------------\nFreeAvatar: Robust 3D Facial Animation Transfer by Learni
 ng an Expression Foundation Model\n\nVideo-driven 3D facial animation tran
 sfer aims to drive avatars to reproduce the expressions of actors. Existin
 g methods have achieved remarkable results by constraining both geometric 
 and perceptual consistency. However, geometric constraints (like those des
 igned on facial landmarks) are insufficie...\n\n\nFeng Qiu and Wei Zhang (
 Netease); Chen Liu (University of Queensland, Netease); Rudong An, Linchen
 g Li, Yu Ding, Changjie Fan, and Zhipeng Hu (Netease); and Xin Yu (Univers
 ity of Queensland)\n---------------------\nHigh-quality Animatable Eyelid 
 Shapes from Lightweight Captures\n\nHigh-quality eyelid reconstruction and
  animation are challenging for the subtle details and complicated deformat
 ions. Previous works usually suffer from the trade-off between the capture
  costs and the quality of details. In this paper, we propose a novel metho
 d that can achieve detailed eyelid recon...\n\n\nJunfeng Lyu and Feng Xu (
 Tsinghua University, China)\n---------------------\nGeometry-Aware Retarge
 ting for Two-Skinned Characters Interaction\n\nInteractive motion between 
 multiple characters is widely utilized in games and movies. However, the m
 ethod for transitioning interactive motions considering the character's di
 verse mesh shape has yet to be studied. We propose a Spatio Cooperative Tr
 ansformer (SCT) to retarget the interacting motions...\n\n\nInseo Jang, So
 ojin Choi, Seokhyeon Hong, Chaelin Kim, and Junyong Noh (Korea Advanced In
 stitute of Science and Technology (KAIST))\n---------------------\nNear-re
 altime Facial Animation by Deep 3D Simulation Super-Resolution\n\nWe prese
 nt a neural network-based simulation super-resolution framework that enhan
 ces facial performance from a low-cost, real-time simulation to a detail l
 evel similar to high-resolution, offline simulators. By training on paired
  low- and high-resolution frames, our approach generalizes to unseen e...\
 n\n\nHyojoon Park and Sangeetha Grama Srinivasan (University of Wisconsin-
 Madison); Matthew Cong, Doyub Kim, Byungsoo Kim, Jonathan Swartz, and Ken 
 Museth (NVIDIA); and Eftychios Sifakis (University of Wisconsin-Madison, N
 VIDIA)\n---------------------\nMotionFix: Text-Driven 3D Human Motion Edit
 ing\n\nThe focus of this paper is 3D motion editing. Given a 3D human moti
 on\nand a textual description of the desired modification, our goal is to 
 generate\nan edited motion as described by the text. The challenges includ
 e the lack\nof training data and the design of a model that faithfully edi
 ts the source\n...\n\n\nNikos Athanasiou (Max Planck Institute for Intelli
 gent Systems); Alpár Cseke (Max Planck Institute for Intelligent Systems, 
 Meshcapade); Markos Diomataris and Michael J. Black (Max Planck Institute 
 for Intelligent Systems); and Gül Varol (LIGM,  ́Ecole des Ponts, Univ Gus
 tave Eiffel, CNRS)\n---------------------\nRefined Inverse Rigging: A Bala
 nced Approach to High-fidelity Blendshape Animation\n\nIn this paper, we p
 resent an advanced approach to solving the inverse rig problem in blendsha
 pe animation, using high-quality corrective blendshapes. Our algorithm int
 roduces novel enhancements in three key areas: ensuring high data fidelity
  in reconstructed meshes, achieving greater sparsity in wei...\n\n\nStevo 
 Racković (University of Lisbon), Dušan Jakovetić (University of Novi Sad),
  and Cláudia Soares (NOVA School of Science and Technology)\n-------------
 --------\nDiffUHaul: A Training-Free Method for Object Dragging in Images\
 n\nText-to-image diffusion models have proven effective for solving many i
 mage editing tasks.\n    However, the seemingly straightforward task of se
 amlessly relocating objects within a scene remains surprisingly challengin
 g. Existing methods addressing this problem often struggle to function rel
 iably in...\n\n\nOmri Avrahami (Hebrew University of Jerusalem), Rinon Gal
  (Tel Aviv University), Gal Chechik (NVIDIA), Ohad Fried (The Interdiscipl
 inary Center Herzliya), Dani Lischinski (Hebrew University of Jerusalem), 
 and Arash Vahdat and Weili Nie (NVIDIA)\n---------------------\nInstantDra
 g: Improving Interactivity in Drag-based Image Editing\n\nDrag-based image
  editing has recently gained popularity for its interactivity and precisio
 n. However, despite the ability of text-to-image models to generate sample
 s within a second, drag editing still lags behind due to the challenge of 
 accurately reflecting user interaction while maintaining image...\n\n\nJoo
 nghyuk Shin (Seoul National University), Daehyeon Choi (POSTECH), and Jaes
 ik Park (Seoul National University)\n---------------------\nConsolidating 
 Attention Features for Multi-view Image Editing\n\nLarge-scale text-to-ima
 ge models enable a wide range of image editing techniques, using text prom
 pts or even spatial controls. However, applying these editing methods to m
 ulti-view images depicting a single scene leads to 3D-inconsistent results
 . In this work, we focus on spatial control-based geome...\n\n\nOr Patashn
 ik (Tel Aviv University); Rinon Gal (Tel Aviv University, NVIDIA Research)
 ; Daniel Cohen-Or (Tel Aviv University); and Jun-Yan Zhu and Fernando De L
 a Torre (Carnegie Mellon University)\n---------------------\nTurboEdit: Te
 xt-Based Image Editing Using Few-Step Diffusion Models\n\nDiffusion models
  have opened the path to a wide range of text-based image editing framewor
 ks. However, these typically build on the multi-step nature of the diffusi
 on backwards process, and adapting them to distilled, fast-sampling method
 s has proven surprisingly challenging. Here, we focus on a pop...\n\n\nGil
 ad Deutch (Tel Aviv University); Rinon Gal (Tel Aviv University, NVIDIA Re
 search); and Daniel Garibi, Or Patashnik, and Daniel Cohen-Or (Tel Aviv Un
 iversity)\n---------------------\nGenerative Portrait Shadow Removal\n\nWe
  introduce a high-fidelity portrait shadow removal model that can effectiv
 ely enhance the image of a portrait by predicting its appearance under dis
 turbing shadows and highlights. Portrait shadow removal is a highly ill-po
 sed problem where multiple plausible solutions can be found based on a sin
 gl...\n\n\nJae Shin Yoon, Zhixin Shu, Mengwei Ren, Xuaner Zhang, Yannick H
 old-Geoffroy, Krishna Kumar Singh, and He Zhang (Adobe Inc.)\n------------
 ---------\nContent-aware Tile Generation using Exterior Boundary Inpaintin
 g\n\nWe present a novel and flexible learning-based method for generating 
 tileable image sets.  Our method goes beyond simple self-tiling, supportin
 g sets of mutually tileable images that exhibit a high degree of diversity
 .  To promote diversity we decouple structure from content by foregoing ex
 plicit co...\n\n\nSam Sartor and Pieter Peers (College of William & Mary)\
 n---------------------\nEVSplitting: An Efficient and Visually Consistent 
 Splitting Algorithm for 3D Gaussian Splatting\n\nThis paper presents EVSpl
 itting, an efficient and visually consistent splitting algorithm for 3D Ga
 ussian Splatting (3DGS). It is designed to make operating 3DGS as easy and
  effective as other 3D explicit representations, readily for industrial pr
 oductions. The challenges of above target are: 1) The...\n\n\nQi-Yuan Feng
 , Geng-Chen Cao, Hao-Xiang Chen, Qun-Ce Xu, and Tai-Jiang Mu (BNRist, Depa
 rtment of Computer Science and Technology, Tsinghua University); Ralph Mar
 tin (School of Computer Science and Informatics, Cardiff University); and 
 Shi-Min Hu (BNRist, Department of Computer Science and Technology, Tsinghu
 a University)\n---------------------\n3DGSR: Implicit Surface Reconstructi
 on with 3D Gaussian Splatting\n\nIn this paper, we present an implicit sur
 face reconstruction method with\n3D Gaussian Splatting (3DGS), namely 3DGS
 R, that allows for accurate 3D\nreconstruction with intricate details whil
 e inheriting the high efficiency and\nrendering quality of 3DGS. The key i
 nsight is to incorporate an implicit\nsig...\n\n\nXiaoyang Lyu, Yang-Tian 
 Sun, Yi-Hua Huang, Xiuzhe Wu, and Ziyi Yang (University of Hong Kong); Yil
 un Chen and Jiangmiao Pang (Shanghai Artificial Intelligence Laboratory); 
 and Xiaojuan Qi (University of Hong Kong)\n---------------------\nGaussian
 Object: High-Quality 3D Object Reconstruction from Four Views with Gaussia
 n Splatting\n\nReconstructing and rendering 3D objects from highly sparse 
 views is of critical importance for promoting applications of 3D vision te
 chniques and improving user experience. However, images from sparse views 
 only contain very limited 3D information, leading to two significant chall
 enges: 1) Difficult...\n\n\nChen Yang and Sikuang Li (Shanghai Jiao Tong U
 niversity), Jiemin Fang (Huawei), Ruofan Liang (University of Toronto), Li
 ngxi Xie and Xiaopeng Zhang (Huawei), Wei Shen (Shanghai Jiao Tong Univers
 ity), and Qi Tian (Huawei)\n---------------------\nReal-time Large-scale D
 eformation of Gaussian Splatting\n\nNeural implicit representations, inclu
 ding Neural Distance Fields and Neural Radiance Fields, have demonstrated 
 significant capabilities for reconstructing surfaces with complicated geom
 etry and topology, and generating novel views of a scene. Nevertheless, it
  is challenging for users to directly de...\n\n\nLin Gao (Institute of Com
 puting Technology, Chinese Academy of Sciences; University of Chinese Acad
 emy of Sciences); Jie Yang (Institute of Computing Technology, Chinese Aca
 demy of Sciences); Bo-Tao Zhang, Jia-Mu Sun, and Yu-Jie Yuan (Institute of
  Computing Technology, Chinese Academy of Sciences; University of Chinese 
 Academy of Sciences); Hongbo Fu (Hong Kong University of Science and Techn
 ology); and Yu-Kun Lai (Cardiff University)\n---------------------\nBlobGE
 N-3D: Compositional 3D-Consistent Freeview Image Generation with 3D Blobs\
 n\nRecent advances in text-to-image diffusion models have significantly en
 hanced image generation quality, when trained on internet-scale data. Howe
 ver, existing methods are constrained by their reliance on image or scene-
 level conditions, limiting their ability to synthesize composable 3D objec
 ts in a...\n\n\nChao Liu, Weili Nie, Sifei Liu, Abhishek Badki, Hang Su, M
 orteza Mardani, Benjamin Eckart, and Arash Vahdat (NVIDIA)\n--------------
 -------\nL3DG: Latent 3D Gaussian Diffusion\n\nWe propose L3DG, the first 
 approach for generative 3D modeling of 3D Gaussians through a latent 3D Ga
 ussian diffusion formulation.\nThis enables effective generative 3D modeli
 ng, scaling to generation of entire room-scale scenes which can be very ef
 ficiently rendered.\nTo enable effective synthesis of...\n\n\nBarbara Roes
 sle (Technical University of Munich); Norman Müller, Lorenzo Porzi, Samuel
  Rota Bulò, and Peter Kontschieder (Meta Reality Labs); and Angela Dai and
  Matthias Nießner (Technical University of Munich)\n---------------------\
 nEnd-to-End Hybrid Refractive-Diffractive Lens Design with Differentiable 
 Ray-Wave Model\n\nHybrid refractive-diffractive lenses combine the light e
 fficiency of refractive lenses with the information encoding power of diff
 ractive optical elements (DOE), showing great potential as the next genera
 tion of imaging systems. However, accurately simulating such hybrid design
 s is generally difficu...\n\n\nXinge Yang and Matheus Souza (King Abdullah
  University of Science and Technology (KAUST)); Kunyi Wang (King Abdullah 
 University of Science and Technology (KAUST), University of British Columb
 ia); Praneeth Chakravarthula (University of North Carolina at Chapel Hill 
 (UNC)); and Qiang Fu and Wolfgang Heidrich (King Abdullah University of Sc
 ience and Technology (KAUST))\n---------------------\nPlug-and-Play Algori
 thms for Dynamic Non-line-of-sight Imaging\n\nNon-line-of-sight (NLOS) ima
 ging has the ability to recover 3D images of scenes outside the direct lin
 e of sight, which is of growing interest for diverse applications. Despite
  the remarkable progress, NLOS imaging of dynamic objects is still challen
 ging. It requires a large amount of multibounce ph...\n\n\nXin Yuan (Westl
 ake University), JunTian Ye and Yu Hong (University of Science and Technol
 ogy of China), Xiongfei Su (Westlake University), and Feihu Xu (University
  of Science and Technology of China)\n---------------------\nCoherent Opti
 cal Modems for Full-Wavefield Lidar\n\nThe advent of the digital age has d
 riven the development of coherent optical modems---devices that modulate t
 he amplitude and phase of light in multiple polarization states. These mod
 ems transmit data through fiber optic cables that are thousands of kilomet
 ers in length at data rates exceeding one t...\n\n\nParsa Mirdehghan (Univ
 ersity of Toronto, Vector Institute); Brandon Buscaino (Ciena Corporation)
 ; Maxx Wu (University of Toronto, Vector Institute); Doug Charlton and Moh
 ammad E. Mousa-Pasandi (Ciena Corporation); and Kiriakos N. Kutulakos and 
 David B. Lindell (University of Toronto, Vector Institute)\n--------------
 -------\nLearned Multi-aperture Color-coded Optics for Snapshot Hyperspect
 ral Imaging\n\nLearned optics, which incorporate lightweight diffractive o
 ptics, coded-aperture modulation, and specialized image-processing neural 
 networks, have recently garnered attention in the field of snapshot hypers
 pectral imaging (HSI). While conventional methods typically rely on a sing
 le lens element pai...\n\n\nZheng Shi (Princeton University); Xiong Dun (T
 ongji University); Haoyu Wei (University of Hong Kong); Siyu Dong, Zhansha
 n Wang, and Xinbin Cheng (Tongji University); Felix Heide (Princeton Unive
 rsity); and Yifan Peng (University of Hong Kong)\n---------------------\nT
 ime-Gated Polarization for Active Non-Line-Of-Sight Imaging\n\nWe propose 
 a novel method to reconstruct non-line-of-sight (NLOS) scenes that combine
 s polarization and time-of-flight light transport measurements. Unpolarize
 d NLOS imaging methods reconstruct objects hidden around corners by invert
 ing time-gated indirect light paths measured at a visible relay sur...\n\n
 \nOscar Pueyo-Ciutad and Julio Marco (Universidad de Zaragoza - I3A); Step
 hane Schertzer, Frank Christnacher, and Martin Laurenzis (French-German Re
 search Institute of Saint-Louis); and Diego Gutierrez and Albert Redo-Sanc
 hez (Universidad de Zaragoza - I3A)\n---------------------\nRobot Motion D
 iffusion Model: Motion Generation for Robotic Characters\n\nRecent advance
 ments in generative motion models have achieved remarkable results, enabli
 ng the synthesis of lifelike human motions from textual descriptions. Thes
 e kinematic approaches, while visually appealing, often produce motions th
 at fail to adhere to physical constraints, resulting in artifact...\n\n\nA
 gon Serifi (ETH Zürich, Disney Research); Ruben Grandia and Espen Knoop (D
 isney Research); Markus Gross (ETH Zürich, Disney Research); and Moritz Bä
 cher (Disney Research)\n---------------------\nPC-Planner: Physics-Constra
 ined Self-Supervised Learning for Robust Neural Motion Planning with Shape
 -Aware Distance Function\n\nMotion Planning (MP) is a critical challenge i
 n robotics, especially pertinent with the burgeoning interest in embodied 
 artificial intelligence. Traditional MP methods often struggle with high-d
 imensional complexities. Recently neural motion planners, particularly phy
 sics-informed neural planners ba...\n\n\nXujie Shen, Haocheng Peng, and Ze
 song Yang (State Key Laboratory of CAD&CG, Zhejiang University); Juzhan Xu
  (Shenzhen University (SZU)); Hujun Bao (State Key Laboratory of CAD&CG, Z
 hejiang University); Ruizhen Hu (Shenzhen University (SZU)); and Zhaopeng 
 Cui (State Key Laboratory of CAD&CG, Zhejiang University)\n---------------
 ------\nA Plentoptic 3D Vision System\n\nWe present a novel multi-camera, 
 multi-modal vision system designed for industrial robotics applications. T
 he system generates high-quality 3D point clouds, with a focus on improvin
 g the completeness and reducing hallucinations for collision avoidance acr
 oss various geometries, materials, and lighti...\n\n\nAgastya Kalra, Vage 
 Tamaazyan, Alberto Dall'olio, Raghav Khanna, Tomas Gerlich, Georgia Gianno
 polou, Guy Stoppi, Daniel Baxter, and Abhijit Ghosh (Intrinsic); Rick Szel
 iski (Google Research); and Kartik Venkataraman (Intrinsic)\n-------------
 --------\nActuators A La Mode: Modal Actuations for Soft Body Locomotion\n
 \nTraditional character animation specializes in characters with a rigidly
  articulated skeleton and a bipedal/quadripedal morphology. This assumptio
 n simplifies many aspects for designing physically based animations, like 
 locomotion, but comes with the price of excluding characters of arbitrary 
 deform...\n\n\nOtman Benchekroun (University of Toronto); Kaixiang Xie (Mc
 Gill University); Hsueh-Ti Derek Liu (Roblox); Eitan Grinspun (University 
 of Toronto); Sheldon Andrews (Ecole de Technologie Superieure, Roblox); an
 d Victor Zordan (Roblox)\n---------------------\nDecoupling Contact for Fi
 ne-Grained Motion Style Transfer\n\nMotion style transfer changes the styl
 e of a motion while retaining its content and is useful in computer animat
 ions and games. Contact is an essential component of motion style transfer
  that should be controlled explicitly in order to express the style vividl
 y while enhancing motion naturalness and...\n\n\nXiangjun Tang and Linjun 
 Wu (State Key Laboratory of CAD&CG, Zhejiang University); He Wang (Univers
 ity College London (UCL)); Yiqian Wu (State Key Laboratory of CAD&CG, Zhej
 iang University); Bo Hu, Songnan Li, Xu Gong, Yuchen Liao, and Qilong Kou 
 (Tencent Technology (Shenzhen) Co., Ltd.); and Xiaogang Jin (State Key Lab
 oratory of CAD&CG, Zhejiang University)\n---------------------\nMaskedMimi
 c: Unified Physics-Based Character Control Through Masked Motion Inpaintin
 g\n\nCrafting a single, versatile physics-based controller that can breath
 e life into interactive characters across a wide spectrum of scenarios rep
 resents an exciting frontier in character animation. An ideal controller s
 hould support diverse control modalities, such as sparse target keyframes,
  text ins...\n\n\nChen Tessler (NVIDIA Research), Yunrong Guo (NVIDIA), Of
 ir Nabati and Gal Chechik (NVIDIA Research), and Xue Bin Peng (NVIDIA)\n--
 -------------------\nGarVerseLOD: High-Fidelity 3D Garment Reconstruction 
 from a Single In-the-Wild Image using a Dataset with Levels of Details\n\n
 Neural implicit functions have brought impressive advances to the state-of
 -the-art of clothed human digitization from multiple or even single images
 . However, despite the progress, current arts still have difficulty genera
 lizing to unseen images with complex cloth deformation and body poses. In 
 this...\n\n\nZhongjin Luo, Haolin Liu, Chenghong Li, Wanghao Du, Zirong Ji
 n, and Wanhu Sun (Chinese University of Hong Kong, Shenzhen); Yinyu Nie (H
 uawei Technologies Ltd.); Weikai Chen (Tencent America); and Xiaoguang Han
  (Chinese University of Hong Kong, Shenzhen)\n---------------------\nFabri
 cDiffusion: High-Fidelity Texture Transfer for 3D Garments Generation from
  In-The-Wild Images\n\nWe introduce FabricDiffusion, a method for transfer
 ring fabric textures from a single clothing image to 3D garments of arbitr
 ary shapes. Existing approaches typically synthesize textures on the garme
 nt surface through 2D-to-3D texture mapping or depth-aware inpainting via 
 generative models. Unfortun...\n\n\nCheng Zhang (Carnegie Mellon Universit
 y, Texas A&M University); Yuanhao Wang and Francisco Vicente (Carnegie Mel
 lon University); Chenglei Wu, Jinlong Yang, and Thabo Beeler (Google Inc.)
 ; and Fernando De la Torre (Carnegie Mellon University)\n-----------------
 ----\nChebyshev Parameterization for Woven Fabric Modeling\n\nDistortion-m
 inimizing surface parameterization is an essential step for computing 2D p
 ieces necessary to fabricate a target 3D shape from flat material. Garment
  design and textile fabrication are a prominent application example. Commo
 n distortion measures quantify length, angle or area preservation ...\n\n\
 nAnnika Oehri (ETH Zürich) and Aviv Segall, Jing Ren, and Olga Sorkine-Hor
 nung (ETH Zurich)\n---------------------\nUFO Instruction Graphs Are Machi
 ne Knittable\n\nProgramming low-level controls for knitting machines is a 
 meticulous, time-consuming task that demands specialized expertise. Recent
 ly, there has been a shift towards automatically generating low-level knit
 ting machine programs from high-level knit representations that describe k
 nit objects in a mor...\n\n\nJenny Lin (Carnegie Mellon University), Yuka 
 Ikarashi (Massachusetts Institute of Technology), Gilbert Bernstein (Unive
 rsity of Washington), and James McCann (Carnegie Mellon University)\n-----
 ----------------\nVolumetric Homogenization for Knitwear Simulation\n\nWe 
 present volumetric homogenization,  a spatially varying homogenization sch
 eme for knitwear simulation. We are motivated by the observation that macr
 o-scale fabric dynamics is strongly correlated with its underlying knittin
 g patterns. Therefore, homogenization towards a single\nmaterial is less e
 ff...\n\n\nChun Yuan, Haoyang Shi, and Lei Lan (University of Utah); Yuxin
 g Qiu (LightSpeed Studios); Cem Yuksel (University of Utah); Huamin Wang (
 Style3D Research); Chenfanfu Jiang (University of California Los Angeles);
  Kui Wu (LightSpeed Studios); and Yin Yang (University of Utah)\n---------
 ------------\nEnd-to-end Optimization of Fluidic Lenses\n\nPrototyping and
  small volume production of custom imaging-grade lenses is difficult and e
 xpensive, especially for more complex aspherical shapes. Fluidic shaping h
 as recently been proposed as a potential solution: It makes use of the ato
 mic level smoothness of interfaces between liquids, where the s...\n\n\nMu
 lun Na, Hector A. Jimenez Romero, Xinge Yang, Jonathan Klein, Dominik L. M
 ichels, and Wolfgang Heidrich (KAUST)\n---------------------\nComputationa
 l Design of Dense Servers for Immersion Cooling\n\nThe growing demands for
  computational power in cloud computing have led to a significant increase
  in the deployment of high-performance servers. The growing power consumpt
 ion of servers and the heat they produce is on track to outpace the capaci
 ty of conventional air cooling systems, necessitating m...\n\n\nMilin Kodn
 ongbua and Zachary Englhardt (University of Washington); Ricardo Bianchini
  and Rodrigo Fonseca (Microsoft); Alvin Lebeck (Duke University); Daniel S
 . Berger (Microsoft, University of Washington); Vikram Iyer (University of
  Washington); Fiodar Kazhamiaka (Microsoft); and Adriana Schulz (Universit
 y of Washington)\n---------------------\nTune-It: Optimizing Wire Reconfig
 uration for Sculpture Manufacturing\n\nWire sculptures are important in bo
 th industrial applications and daily life. We introduce a novel fabricatio
 n strategy for wire sculptures with complex geometries by tuning the targe
 t shape to a collision-free shape for the wire-bending machine and then be
 nding it back to the target by a human. The...\n\n\nQibing Wu, Zhihao Zhan
 g, Xin Yan, and Fanchao Zhong (Shandong University); Yueze Zhu (University
  College London (UCL)); and Xurong Lu, Runze Xue, Rui Li, Changhe Tu, and 
 Haisen Zhao (Shandong University)\n---------------------\nMotion-Driven Ne
 ural Optimizer for Prophylactic Braces Made by Distributed Microstructures
 \n\nJoint injuries, and their long-term consequences, present a substantia
 l global health burden. Wearable prophylactic braces are an attractive pot
 ential solution to reduce the incidence of joint injuries by limiting join
 t movements that are related to injury risk. Given human motion and ground
  reactio...\n\n\nXingjian Han (Boston University, University of Manchester
 ); Yu Jiang (Dalian University of Technology, University of Manchester); W
 eiming Wang (University of Manchester); Guoxin Fang (Chinese University of
  Hong Kong); Simeon Gill (University of Manchester); Zhiqiang Zhang (Unive
 rsity of Leeds); Shengfa Wang (Dalian University of Technology); Jun Saito
  (Adobe Research); Deepak Kumar (Boston University); Zhongxuan Luo (Dalian
  University of Technology); Emily Whiting (Boston University); and Charlie
  Wang (University of Manchester)\n---------------------\nmpcMech: Multi-Po
 int Conjugation Mechanisms\n\nA mechanism is an assembly of moving parts i
 nterconnected by joints to transfer an input motion to a desired output mo
 tion. Traditionally, to generate a complex motion, mechanisms are modeled 
 by selecting and combining a number of mechanical parts with simple shapes
  such as links, gears, and cams. C...\n\n\nKe Chen (University of Science 
 and Technology of China), Siqi Li and Peng Song (Singapore University of T
 echnology and Design (SUTD)), Jianmin Zheng (Nanyang Technological Univers
 ity (NTU)), and Ligang Liu (University of Science and Technology of China)
 \n---------------------\nFragmentDiff: A Diffusion Model for Fractured Obj
 ect Assembly\n\nFractured object reassembly is a challenging problem in co
 mputer vision and graphics with applications in industrial manufacturing a
 nd archaeology. Traditional methods based on shape descriptors and geometr
 ic registration often struggle with ambiguous features, resulting in lower
  accuracy. To addres...\n\n\nQun-Ce Xu and Hao-Xiang Chen (BNRist, Departm
 ent of Computer Science and Technology, Tsinghua University); Jiacheng Hua
  (Department of Computer Science and Technology, Tsinghua University); Xia
 ohua Zhan (Department of Foreign Languages and Literatures, Tsinghua Unive
 rsity); Yong-Liang Yang (Department of Computer Science, University of Bat
 h); and Tai-Jiang Mu (BNRist, Department of Computer Science and Technolog
 y, Tsinghua University)\n---------------------\nInstanceTex: Instance-leve
 l Controllable Texture Synthesis for 3D Scenes via Diffusion Priors\n\nAut
 omatically generating high-fidelity texture for a complex scene remains an
  open problem in computer graphics. While pioneering text-to-texture works
  based on 2D diffusion models have achieved fascinating results on single 
 objects, they either suffer from style inconsistency and semantic misalign
 m...\n\n\nMingxin Yang (Shenzhen Institute of Advanced Technology, Chinese
  Academy of Sciences); Jianwei Guo (Institute of Automation, Chinese Acade
 my Of Sciences); Yuzhi Chen (School of Artificial Intelligence, University
  of Chinese Academy of Sciences); Lan Chen (Institute of Automation, Chine
 se Academy of Sciences); Pu Li (Institute of Automation, Chinese Academy O
 f Sciences); Zhanglin Cheng (Shenzhen Institute of Advanced Technology, Ch
 inese Academy of Sciences); Xiaopeng Zhang (Institute of Automation, Chine
 se Academy Of Sciences); and Hui Huang (Shenzhen University (SZU))\n------
 ---------------\nText-Guided Texturing by Synchronized Multi-View Diffusio
 n\n\nThis paper introduces a novel approach to synthesize texture to dress
  up a given 3D object, given a text prompt. \nBased on the pretrained text
 -to-image (T2I) diffusion model, existing methods usually employ a project
 -and-inpaint approach, in which a view of the given object is first genera
 ted and wa...\n\n\nYuxin Liu and Minshan Xie (Chinese University of Hong K
 ong); Hanyuan Liu (City University of Hong Kong); and Tien-Tsin Wong (Mona
 sh University, Chinese University of Hong Kong)\n---------------------\nSt
 yleTex: Style Image-Guided Texture Generation for 3D Models\n\nStyle-guide
 d texture generation aims to generate a texture that is harmonious with bo
 th the style of the reference image and the geometry of the input mesh, gi
 ven a reference style image and a 3D mesh with its text description.  \nAl
 though diffusion-based 3D texture generation methods, such as distil...\n\
 n\nZhiyu Xie, Yuqing Zhang, Xiangjun Tang, Yiqian Wu, and Dehan Chen (Stat
 e Key Laboratory of CAD&CG, Zhejiang University); Gongsheng Li (Zhejiang U
 niversity); and Xiaogang Jin (State Key Laboratory of CAD&CG, Zhejiang Uni
 versity)\n---------------------\nTEXGen: a Generative Diffusion Model for 
 Mesh Textures\n\nWhile high-quality texture maps are essential for realist
 ic 3D asset rendering, few studies have explored learning directly in the 
 texture space, especially on large-scale datasets. In this work, we depart
  from the conventional approach of relying on pre-trained 2D diffusion mod
 els for test-time opt...\n\n\nXin Yu (University of Hong Kong); Ze Yuan (B
 eihang University); Yuan-Chen Guo (VAST); Ying-Tian Liu (Tsinghua Universi
 ty); Jianhui Liu (University of Hong Kong); Yangguang Li, Yan-Pei Cao, and
  Ding Liang (VAST); and Xiaojuan Qi (University of Hong Kong)\n-----------
 ----------\nCompositional Neural Textures\n\nTexture plays a vital role in
  enhancing visual richness in both real photographs and computer-generated
  imagery. However, the process of editing textures often involves laboriou
 s and repetitive manual adjustments of textons, which are the recurring lo
 cal patterns that characterize textures. This wor...\n\n\nPeihan Tu (Unive
 rsity of Maryland, College Park); Li-Yi Wei (Adobe Research); and Matthias
  Zwicker (University of Maryland, College Park)\n---------------------\nCa
 mera Settings as Tokens: Modeling Photography on Latent Diffusion Models\n
 \nText-to-image models have revolutionized content creation, enabling user
 s to generate images from natural language prompts. While recent advanceme
 nts in conditioning these models offer more control over the generated res
 ults, photography—a significant artistic domain—remains inadequately...\n\
 n\nI-Sheng Fang, Yue-Hua Han, and Jun-Cheng Chen (Academia Sinica)\n------
 ---------------\nTencers: Tension-Constrained Elastic Rods\n\nWe study ens
 embles of elastic rods that are tensioned by a small set of inextensible c
 ables. The cables induce forces that deform the initially straight, but fl
 exible rods into 3D space curves at equilibrium. Rods can be open or close
 d, knotted, and arranged in arbitrary topologies. We specifically ...\n\n\
 nLiliane-Joy Dandy, Michele Vidulis, Yingying Ren, and Mark Pauly (EPFL)\n
 ---------------------\nQ3T Prisms: A Linear-Quadratic Solid Shell Element 
 for Elastoplastic Surfaces\n\nWe introduce a novel approach for simulating
  elastoplastic surfaces using quadratic through-the-thickness (Q3T) solid 
 shell elements. Modeling the mechanics of deformable surfaces has been a c
 ornerstone of graphics research for decades. Although thin shell models ar
 e suitable for many materials and ...\n\n\nJuan Sebastian Montes Maestre, 
 Stelian Coros, and Bernhard Thomaszewski (ETH Zürich)\n-------------------
 --\nA Mesh-based Simulation Framework using Automatic Code Generation\n\nO
 ptimized parallel implementations on GPU or CPU have dramatically enhanced
  the fidelity, resolution and accuracy of physical simulations and mesh-ba
 sed algorithms. However, attaining optimal performance requires expert kno
 wledge and might demand complex code and memory layout optimizations. This
  ad...\n\n\nPhilipp Herholz, Tuur Stuyck, and Ladislav Kavan (Meta)\n-----
 ----------------\nPolar Interpolants for Thin-Shell Microstructure Homogen
 ization\n\nThis paper introduces a new formulation for material homogeniza
 tion of thin-shell microstructures. It addresses important challenges that
  limit the quality of previous approaches: methods that fit the energy res
 ponse neglect visual impact, methods that fit the stress response are not 
 conservative, a...\n\n\nAntoine Chan-Lock (Universidad Rey Juan Carlos) an
 d Miguel A. Otaduy (Universidad Rey Juan Carlos, Facebook Reality Labs)\n-
 --------------------\nXPBI: Position-Based Dynamics with Smoothing Kernels
  Handles Continuum Inelasticity\n\nPBD and its extension, XPBD, have been 
 predominantly applied to compliant constrained elastodynamics, with their 
 potential in finite strain (visco-) elastoplasticity remaining underexplor
 ed. XPBD is often perceived to stand in contrast to other meshless methods
 , such as the MPM. MPM is based on disc...\n\n\nChang Yu and Xuan Li (Univ
 ersity of California Los Angeles), Lei Lan and Yin Yang (University of Uta
 h), and Chenfanfu Jiang (University of California Los Angeles)\n----------
 -----------\nMulti-Resolution Real-Time Deep Pose-Space Deformation\n\nWe 
 present a hard-real-time multi-resolution mesh shape deformation technique
  for skeleton-driven soft-body characters. Producing mesh deformations at 
 multiple levels of detail is very important in many applications in comput
 er graphics. Our work targets applications where the multi-resolution shap
 e...\n\n\nMianlun Zheng and Jernej Barbic (University of Southern Californ
 ia)\n---------------------\nA Statistical Approach to Monte Carlo Denoisin
 g\n\nThe stochastic nature of modern Monte Carlo (MC) rendering methods in
 evitably produces noise in rendered images for a practical number of sampl
 es per pixel. The problem of denoising these images has been widely studie
 d, with most recent methods relying on data-driven, pretrained neural netw
 orks. In ...\n\n\nHiroyuki Sakai and Christian Freude (Technical Universit
 y of Vienna), Thomas Auzinger (Institute of Science and Technology Austria
 ), and David Hahn and Michael Wimmer (Technical University of Vienna)\n---
 ------------------\nOnline Neural Denoising with Cross-Regression for Inte
 ractive Rendering\n\nGenerating a rendered image sequence through Monte Ca
 rlo ray tracing is an appealing option when one aims to accurately simulat
 e various lighting effects. Unfortunately, interactive rendering scenarios
  limit the allowable sample size for such sampling-based light transport a
 lgorithms, resulting in a...\n\n\nHajin Choi (Gwangju Institute of Science
  and Technology); Seokpyo Hong (Samsung Advanced Institute of Technology);
  Inwoo Ha (Samsung Advanced Institute of Technology, KAIST); Nahyup Kang (
 Samsung Advanced Institute of Technology); and Bochang Moon (Gwangju Insti
 tute of Science and Technology)\n---------------------\nFiltering-Based Re
 construction for Gradient-Domain Rendering\n\nGradient-domain rendering me
 thods reconstruct color images based on the Poisson equation with gradient
 s from correlated sampling. The relatively low variance in the gradient es
 timation facilitates convergence but the inevitable noises make the solvin
 g process prone to unpleasant spiky artifacts.\n\nWe...\n\n\nDifei Yan and
  Shaokun Zheng (Tsinghua University), Ling-Qi Yan (University of Californi
 a Santa Barbara), and Kun Xu (Tsinghua University)\n---------------------\
 nSpatiotemporal Bilateral Gradient Filtering for Inverse Rendering\n\nIn i
 nverse rendering, gradient-based methods, which have seen great progress i
 n the recent years, are typically used in conjunction with the Adam optimi
 zer. While Adam usually improves convergence by temporally filtering gradi
 ents over previous iterations to reduce noise, it is not tailored to inver
 ...\n\n\nWesley Chang, Xuanda Yang, Yash Belhe, Ravi Ramamoorthi, and Tzu-
 Mao Li (University of California San Diego)\n---------------------\nNeural
  Kernel Regression for Consistent Monte Carlo Denoising\n\nUnbiased Monte 
 Carlo path tracing that is extensively used in realistic rendering produce
 s undesirable noise, especially with low samples per pixel (spp). Recently
 , several methods have coped with this problem by importing unbiased noisy
  images and auxiliary features to neural networks to either pre...\n\n\nQi
  Wang (State Key Laboratory of CAD&CG, Zhejiang University); Pengju Qiao (
 Institute of Software Chinese Academy of Sciences; State Key Laboratory of
  CAD&CG, Zhejiang University); Yuchi Huo (State Key Laboratory of CAD&CG, 
 Zhejiang University; Zhejiang University); Shiji Zhai (Institute of Comput
 ing Technology, Chinese Academy of Sciences); Zixuan Xie (Institute of Com
 puting Technology, Chinese Academy of Sciences; Zhejiang Lab); Rengan Xie 
 (State Key Laboratory of CAD&CG, Zhejiang University); Wei Hua (Zhejiang L
 ab); Hujun Bao (State Key Laboratory of CAD&CG, Zhejiang University; Zheji
 ang University); and Tao Liu (Shanghai Maritime University, College of Tra
 nsport & Communications)\n---------------------\nDynamic Skeletonization v
 ia Variational Medial Axis Sampling\n\nWe present a novel method for compu
 ting a discrete skeleton from a shape represented by a point cloud or tria
 ngle mesh.\nInspired by variational shape approximation, our approach opti
 mizes the partitioning of the input shape by minimizing an error metric de
 fined between medial axis samples (medial sp...\n\n\nQijia HUANG, Pierre K
 RAEMER, Sylvain THERY, and Dominique BECHMANN (Université de Strasbourg; I
 Cube, CNRS)\n---------------------\nMedial Skeletal Diagram: A Generalized
  Medial Axis Approach for 3D Shape Representation\n\nWe propose the Medial
  Skeletal Diagram, a novel skeletal representation that tackles the prevai
 ling issues around skeleton sparsity and reconstruction accuracy in existi
 ng skeletal representations. Our approach augments the continuous elements
  in the medial axis representation to effectively shift t...\n\n\nMinghao 
 Guo, Bohan Wang, and Wojciech Matusik (MIT CSAIL)\n---------------------\n
 MATTopo: Topology-preserving Medial Axis Transform with Restricted Power D
 iagram\n\nWe present a novel topology-preserving 3D medial axis computatio
 n framework based on volumetric restricted power diagram (RPD), while pres
 erving the medial features and geometric convergence simultaneously, for b
 oth 3D CAD and organic shapes. The volumetric RPD discretizes the input 3D
  volume into s...\n\n\nNingna Wang (University of Texas at Dallas), Hui Hu
 ang (Shenzhen University (SZU)), Shibo Song (Independent Researcher), Bin 
 Wang (Tsinghua University), Wenping Wang (Texas A&M University), and Xiaoh
 u Guo (University of Texas at Dallas)\n---------------------\nDeformation 
 Recovery: Localized Learning for Detail-Preserving Deformations\n\nWe intr
 oduce a novel data-driven approach aimed at designing high-quality shape d
 eformations based on a coarse localized input signal. Unlike previous data
 -driven methods that require a global shape encoding, we observe that deta
 il-preserving deformations can be estimated reliably without any global...
 \n\n\nRamana Sundararaman (Centre National de la Recherche Scientifique - 
 Laboratoire d'informatique de l'École Polytechnique (LIX)); Nicolas Donati
  (Ansys); Simone Melzi (University of Milano-Bicocca); Etienne Corman (Uni
 versité de Lorraine, CNRS); and Maks Ovsjanikov (Centre National de la Rec
 herche Scientifique - Laboratoire d'informatique de l'École Polytechnique 
 (LIX))\n---------------------\nPolynomial Cauchy Coordinates for Curved Ca
 ges\n\nBarycentric coordinates are widely used in computer graphics, espec
 ially in shape deformation. Traditionally, barycentric coordinates are def
 ined for polygonal domains. In this work, we relax this requirement by rep
 resenting the boundary of the domain using a Bézier spline and extend the 
 complex-val...\n\n\nZhehui Lin and Renjie Chen (University of Science and 
 Technology of China)\n---------------------\nC^0 Generalized Coons Patches
  for High-order Cage-based Deformation\n\nSpace deformations deform the am
 bient space and thus implicitly deform the embedded objects. Free-Form Def
 ormation allows high-order deformation to the embedding space, yet the lat
 tice may fail to conform to the object and involves many internal control 
 points. Cage-based Deformation utilizes a cage...\n\n\nKaikai Qin, Yunhao 
 Zhou, Chenhao Ying, Yajuan Li, and Chongyang Deng (Hangzhou Dianzi Univers
 ity)\n---------------------\nA Time-Dependent Inclusion-Based Method for C
 ontinuous Collision Detection between Parametric Surfaces\n\nContinuous co
 llision detection (CCD) between parametric surfaces is typically formulate
 d as a five-dimensional constrained optimization problem. In the field of 
 CAD and computer graphics, common approaches to solving this problem rely 
 on linearization or sampling strategies. Alternatively, inclusion...\n\n\n
 Xuwen Chen and Cheng Yu (School of Intelligence Science and Technology, Pe
 king University; State Key Laboratory of General Artificial Intelligence);
  Xingyu Ni (School of Computer Science, Peking University; State Key Labor
 atory of General Artificial Intelligence); Mengyu Chu (School of Intellige
 nce Science and Technology, Peking University; State Key Laboratory of Gen
 eral Artificial Intelligence); Bin Wang (Beijing Institute for General Art
 ificial Intelligence (BIGAI), State Key Laboratory of General Artificial I
 ntelligence); and Baoquan Chen (School of Intelligence Science and Technol
 ogy, Peking University; State Key Laboratory of General Artificial Intelli
 gence)\n---------------------\nA Cubic Barrier with Elasticity-Inclusive D
 ynamic Stiffness\n\nThis paper presents a new cubic barrier with elasticit
 y-inclusive dynamic stiffness for penetration-free contact resolution and 
 strain limiting. We show that our method enlarges tight strain-limiting ga
 ps where logarithmic barriers struggle and enables highly scalable contact
 -rich simulation.\n\n\nRyoichi Ando (ZOZO)\n---------------------\nBarrier
 -Augmented Lagrangian for GPU-based Elastodynamic Contact\n\nWe propose a 
 GPU-based iterative method for accelerated elastodynamic simulation with t
 he log-barrier-based contact model. While Newton's method is a conventiona
 l choice for solving the interior-point system, the presence of ill-condit
 ioned log barriers often necessitates a direct solution at each l...\n\n\n
 Dewen Guo (Peking University), Minchen Li (Carnegie Mellon University), Yi
 n Yang (University of Utah), and Sheng Li and Guoping Wang (Peking Univers
 ity)\n---------------------\nEfficient GPU Cloth Simulation with Non-dista
 nce Barriers and Subspace Reuse\n\nThis paper pushes the performance of cl
 oth simulation, making the simulation interactive even for high-resolution
  garment models while keeping every triangle untangled. The penetration-fr
 ee guarantee is inspired by the interior point method, which converts the 
 inequality constraints to barrier poten...\n\n\nLei Lan, Zixuan Lu, Jingyi
  Long, and Chun Yuan (University of Utah); Xuan Li (University of Californ
 ia Los Angeles); Xiaowei He (Institute of software, Chinese Academy of Sci
 ences); Huamin Wang (Style3D Research); Chenfanfu Jiang (University of Cal
 ifornia Los Angeles); and Yin Yang (University of Utah)\n-----------------
 ----\ngDist: Efficient Distance Computation between 3D Meshes on GPU\n\nCo
 mputing maximum/minimum distances between 3D meshes is crucial for various
  applications, i.e., robotics, CAD, VR/AR, etc. In this work, we introduce
  a highly parallel algorithm (gDist) optimized for Graphics Processing Uni
 ts (GPUs), which is capable of computing the distance between two meshes w
 it...\n\n\nPeng Fan, Wei Wang, and Ruofeng Tong (Zhejiang University); Hai
 long Li (Poisson Soft); and Min Tang (Zhejiang University, Zhejiang Sci-Te
 ch University)\n---------------------\nTrading Spaces: Adaptive Subspace T
 ime Integration for Contacting Elastodynamics\n\nWe construct a subspace s
 imulator that adaptively balances solution improvement against system size
 . The core components of our simulator are an adaptive subspace oracle, mo
 del, and parallel time-step solver algorithm. Our in-time-step adaptivity 
 oracle continually assesses subspace solution quality...\n\n\nTy Trusty (U
 niversity of Toronto); Yun (Raymond) Fei (Adobe Research); David Levin (Un
 iversity of Toronto, NVIDIA Research); and Danny Kaufman (Adobe Research)\
 n---------------------\nInverse Rendering for Tomographic Volumetric Addit
 ive Manufacturing\n\nTomographic Volumetric Additive Manufacturing (TVAM) 
 is an emerging 3D printing technology that can create complex objects in u
 nder a minute. The key idea is to project intense light patterns onto a ro
 tating vial of photo-sensitive resin, causing polymerization where the cum
 ulative dose\nof these pat...\n\n\nBaptiste Nicolet, Felix Wechsler, Jorge
  Madrid-Wolff, Christophe Moser, and Wenzel Jakob (EPFL)\n----------------
 -----\nLearning Based Toolpath Planner on Diverse Graphs for 3D Printing\n
 \nThis paper presents a learning based planner for computing optimized 3D 
 printing toolpaths on prescribed graphs, the challenges of which include t
 he varying graph structures on different models and the large scale of nod
 es & edges on a graph. We adopt an on-the-fly strategy to tackle these cha
 llenge...\n\n\nYuming Huang, Yuhu Guo, and Renbo Su (University of Manches
 ter); Xingjian Han (Boston University); Junhao Ding (Chinese University of
  Hong Kong); Tianyu Zhang, Tao Liu, and Weiming Wang (University of Manche
 ster); Guoxin Fang and Xu Song (Chinese University of Hong Kong); Emily Wh
 iting (Boston University); and Charlie Wang (University of Manchester)\n--
 -------------------\nDifferentiable Modeling of Material Spreading in Inkj
 et Printing for Appearance Prediction\n\nInkjet 3D printers produce solid 
 shapes using very small voxels made of polymeric materials. While state-of
 -the-art methods for predicting the appearance of inkjet-printed objects a
 ssume a perfect grid, the printed patterns have an irregular material dist
 ribution due to complex spreading behavior. T...\n\n\nEmiliano Luci (Max P
 lanck Institute for Informatics), Fabio Pellacini (Università degli Studi 
 di Modena e Reggio Emilia), and Vahid Babaei (Max Planck Institute for Inf
 ormatics)\n---------------------\nComputational Design of a Kit of Parts f
 or Bending Active Structures\n\nBending-active structures are composed of 
 elastic elements that deform to achieve a desired target shape. To support
  effective design, inverse algorithms have been proposed that optimize the
  geometry of each element specifically for each design. This makes it diff
 icult to reuse elements across desig...\n\n\nQuentin Becker, Uday Kusupati
 , Seiichi Suzuki, and Mark Pauly (EPFL)\n---------------------\nA Flexible
  Mold for Facade Panel Fabrication\n\nArchitectural surface panelling ofte
 n requires fabricating molds for panels, a process that can be cost-ineffi
 cient and material-wasteful when using traditional methods such as CNC mil
 ling. In this paper, we introduce a novel solution to generating molds for
  efficiently fabricating architectural pan...\n\n\nFlorian Rist (KAUST, TU
  Wien); Zhecheng Wang (UofT); Davide Pellis (ISTI-CNR); Marco Palma (TU Wi
 en); Daoming Liu (KAUST); Eitan Grinspun (UofT); and Dominik L. Michels (K
 AUST)\n---------------------\nMillimetric Human Surface Capture in Minutes
 \n\nDetailed human surface capture from multiple images is an essential co
 mponent for many 3D production, analysis and transmission tasks. Yet produ
 cing millimetric precision 3D models in practical time, and actually verif
 ying their 3D accuracy in a real-world capture context, remain key challen
 ges due ...\n\n\nBriac Toussaint and Laurence Boissieux (Centre Inria de l
 ’Université Grenoble Alpes); Diego Thomas (Kyushu University); Edmond Boye
 r (Meta Reality Labs Research); and Jean-Sébastien Franco (LJK, CNRS, Gren
 oble INP, Université Grenoble Alpes; Centre Inria de l’Université Grenoble
  Alpes)\n---------------------\nRoMo: A Robust Solver for Full-body Unlabe
 led Optical Motion Capture\n\nOptical motion capture (MoCap) is the "gold 
 standard" for accurately capturing full-body motions. To make use of raw M
 oCap point data, the system labels the points with corresponding body part
  locations and solves the full-body motions. However, MoCap data often con
 tains mislabeling, occlusion and p...\n\n\nXiaoyu Pan and Bowen Zheng (Sta
 te Key Laboratory of CAD&CG, Zhejiang University); Xinwei Jiang, Zijiao Ze
 ng, and Qilong Kou (Tencent Games Digital Content Technology Center); He W
 ang (Department of Computer Science and UCL Centre for Artificial Intellig
 ence, University College London); and Xiaogang Jin (State Key Laboratory o
 f CAD&CG, Zhejiang University)\n---------------------\nFürElise: Capturing
  and Physically Synthesizing Hand Motion of Piano Performance\n\nPiano pla
 ying requires agile, precise, and coordinated hand control that stretches 
 the limits of dexterity. Hand motion models with the sophistication to acc
 urately recreate piano playing have a wide range of applications in charac
 ter animation, embodied AI, biomechanics, and VR/AR. In this paper, w...\n
 \n\nRuocheng Wang, Pei Xu, Haochen Shi, Elizabeth Schumann, and C. Karen L
 iu (Stanford University)\n---------------------\nLook Ma, no markers: holi
 stic performance capture without the hassle\n\nWe tackle the problem of hi
 ghly-accurate, holistic performance capture for the face, body and hands s
 imultaneously. Motion-capture technologies used in film and game productio
 n typically focus only on face, body or hand capture independently, involv
 e complex and expensive hardware and a high degree ...\n\n\nCharlie Hewitt
 , Fatemeh Saleh, Sadegh Aliakbarian, Lohit Petikam, Shideh Rezaeifar, Loui
 s Florentin, Zafiirah Hosenie, Thomas J. Cashman, and Julien Valentin (Mic
 rosoft); Darren Cosker (Microsoft, University of Bath); and Tadas Baltrusa
 itis (Microsoft)\n---------------------\nEgoHDM: An Online Egocentric-Iner
 tial Human Motion Capture, Localization, and Dense Mapping System\n\nWe pr
 esent EgoHDM, an online egocentric-inertial human motion capture (mocap), 
 localization, and dense mapping system. Our system uses 6 inertial measure
 ment units (IMUs) and a commodity head-mounted RGB camera. EgoHDM is the f
 irst human mocap system that offers dense scene mapping in near real-time.
 ..\n\n\nHandi Yin and Bonan Liu (Hong Kong University of Science and Techn
 ology, Guangzhou); Manuel Kaufmann (ETH Zürich); Jinhao He (Hong Kong Univ
 ersity of Science and Technology, Guangzhou); Sammy Christen (ETH Zürich);
  and Jie Song and Pan Hui (Hong Kong University of Science and Technology,
  Guangzhou; Hong Kong University of Science and Technology)\n-------------
 --------\nELMO: Enhanced Real-time LiDAR Motion Capture through Upsampling
 \n\nThis paper introduces ELMO, a real-time upsampling motion capture fram
 ework designed for a single LiDAR sensor. Modeled as a conditional autoreg
 ressive transformer-based upsampling motion generator, ELMO achieves 60 fp
 s motion capture from a 20 fps LiDAR point cloud sequence. The key feature
  of ELMO...\n\n\nDeok-Kyeong Jang (MOVIN Inc.); Dongseok Yang (MOVIN Inc.,
  KAIST); Deok-Yun Jang (MOVIN Inc., GIST); Byeoli Choi (MOVIN Inc., KAIST)
 ; Donghoon Shin (MOVIN Inc.); and Sung-Hee Lee (KAIST)\n------------------
 ---\nAdR-Gaussian: Accelerating Gaussian Splatting with Adaptive Radius\n\
 n3D Gaussian Splatting (3DGS) is a recent explicit 3D representation that 
 has achieved high-quality reconstruction and real-time rendering of comple
 x scenes. However, the rasterization pipeline still suffers from unnecessa
 ry overhead resulting from avoidable serial Gaussian culling, and uneven l
 oad d...\n\n\nXinzhe Wang, Ran Yi, and Lizhuang Ma (Shanghai Jiao Tong Uni
 versity)\n---------------------\n3D Gaussian Ray Tracing: Fast Tracing of 
 Particle Scenes\n\nParticle-based representations of radiance fields such 
 as 3D Gaussian Splatting have found great success for reconstructing and r
 e-rendering of complex scenes.\nMost existing methods render particles via
  rasterization, projecting them to screen space tiles for processing in a 
 sorted order.\nThis work ...\n\n\nNicolas Moenne-Loccoz (NVIDIA); Ashkan M
 irzaei (NVIDIA, University of Toronto); and Or Perel, Riccardo de Lutio, J
 anick Martinez Esturo, Gavriel State, Sanja Fidler, Nicholas Sharp, and Za
 n Gojcic (NVIDIA)\n---------------------\nHigh-Throughput Batch Rendering 
 for Embodied AI\n\nIn this paper we study the problem of efficiently rende
 ring images for embodied AI training workloads, where agent training invol
 ves rendering millions to billions of independent frames, often at low-res
 olutions and with simple (or no) lighting and shading, that serve as the a
 gent's observations of ...\n\n\nLuc Guy Rosenzweig, Brennan Shacklett, War
 ren Xia, and Kayvon Fatahalian (Stanford University)\n--------------------
 -\nEfficient Image-Space Shape Splatting for Monte Carlo Rendering\n\nA ty
 pical Monte Carlo rendering method contributes one light path only to a si
 ngle pixel at a time. Reusing light paths across multiple pixels, however,
  can amortize the cost and improve the efficiency. The state of the art of
  path reuse is to employ shift mapping to reduce the cost of path reuse, w
 ...\n\n\nXiaochun Tong and Toshiya Hachisuka (University of Waterloo)\n---
 ------------------\nDirectL: Efficient Radiance Fields Rendering for 3D Li
 ght Field Displays\n\nAutostereoscopic display technology, despite decades
  of development, has not achieved extensive application, primarily due to 
 the daunting challenge of three-dimensional (3D) content creation for non-
 specialists. The emergence of Radiance Field as an innovative 3D represent
 ation has markedly revolut...\n\n\nZongyuan Yang, Baolin Liu, Yingde Song,
  Lan Yi, Yongping Xiong, Zhaohe Zhang, and Xunbo Yu (Beijing University of
  Posts and Telecommunications)\n---------------------\nPDP: Physics-Based 
 Character Animation via Diffusion Policy\n\nGenerating diverse and realist
 ic human motion that can physically interact with an environment remains a
  challenging research area in character animation. Meanwhile, diffusion-ba
 sed methods, as proposed by the robotics community, have demonstrated the 
 ability to capture highly diverse and multi-moda...\n\n\nTakara Truong, Mi
 chael Piseno, Zhaoming Xie, and Karen Liu (Stanford University)\n---------
 ------------\nMonkey See, Monkey Do: Harnessing Self-attention in Motion D
 iffusion for Zero-shot Motion Transfer\n\nGiven the remarkable results of 
 motion synthesis with diffusion models, a natural question arises: how can
  we effectively leverage these models for motion editing? Existing diffusi
 on-based motion editing methods overlook the profound potential of the pri
 or embedded within the weights of pre-trained ...\n\n\nSigal Raab, Inbar G
 at, Nathan Sala, Guy Tevet, and Rotem Shalev-Arkushin (Tel Aviv University
 ); Ohad Fried (Reichman University); and Amit Haim Bermano and Daniel Cohe
 n-Or (Tel Aviv University)\n---------------------\nCBIL: Collective Behavi
 or Imitation Learning for Fish from Real Videos\n\nReproducing realistic c
 ollective behaviors presents a captivating yet formidable challenge. Tradi
 tional rule-based methods rely on hand-crafted principles, limiting motion
  diversity and realism in generated collective behaviors. Recent imitation
  learning methods learn from data but often require gro...\n\n\nYifan Wu (
 University of Hong Kong); Zhiyang Dou (University of Hong Kong, University
  of Pennsylvania); Yuko Ishiwaka and Shun Ogawa (SoftBank); Yuke Lou (Univ
 ersity of Hong Kong); Wenping Wang (Texas A&M University); Lingjie Liu (Un
 iversity of Pennsylvania); and Taku Komura (University of Hong Kong)\n----
 -----------------\nResolving Collisions in Dense 3D Crowd Animations\n\nWe
  propose a contact-aware method for synthesizing dense 3D crowds of animat
 ed characters. Unlike existing methods, our approach prevents character in
 tersections by modeling contacts using physics-based techniques. This resu
 lts in real-time, collision-free animations with enhanced realism and geom
 et...\n\n\nGonzalo Gomez-Nogales, Melania Prieto-Martin, Cristian Romero, 
 Marc Comino-Trinidad, and Pablo Ramon-Prieto (Universidad Rey Juan Carlos)
 ; Anne-Hélène Olivier (INRIA, Université de Rennes, CNRS, IRISA,  M2S Cent
 re de Rennes); Ludovic Hoyet (Institut national de recherche en informatiq
 ue et en automatique (INRIA)); Miguel Otaduy (Universidad Rey Juan Carlos)
 ; Julien Pettre (Institut national de recherche en informatique et en auto
 matique (INRIA)); and Dan Casas (Universidad Rey Juan Carlos)\n-----------
 ----------\nBody Gesture Generation for Multimodal Conversational Agents\n
 \nCreating intelligent virtual agents with realistic conversational abilit
 ies necessitates a multimodal communication approach extending beyond text
 . Body gestures, in particular, play a pivotal role in delivering a lifeli
 ke user experience by providing additional context, such as agreement, con
 fusion...\n\n\nSunwoo Kim, Minwook Chang, and Yoonhee Kim (NCSOFT) and Jeh
 ee Lee (Seoul National University)\n---------------------\nSpaceMesh: A Co
 ntinuous Representation for Learning Manifold Surface Meshes\n\nMeshes are
  ubiquitous in visual computing and simulation, yet most existing machine 
 learning techniques represent meshes only indirectly, e.g. as the level se
 t of a scalar field, or deformation of a template, or as a disordered tria
 ngle soup lacking local structure. This work presents a scheme to di...\n\
 n\nTianchang Shen (University of Toronto, NVIDIA Research) and Zhaoshuo Li
 , Marc Law, Matan Atzmon, Sanja Fidler, James Lucas, Jun Gao, and Nicholas
  Sharp (NVIDIA Research)\n---------------------\nNASM: Neural Anisotropic 
 Surface Meshing\n\nThis paper introduces a new learning-based method, NASM
 , for anisotropic surface meshing. Our key idea is to propose a graph neur
 al network to embed an input mesh into a high-dimensional (high-d) Euclide
 an embedding space to preserve curvature-based anisotropic metric by using
  a dot product loss bet...\n\n\nHongbo Li, Haikuan Zhu, and Sikai Zhong (W
 ayne State University); Ningna Wang (University of Texas at Dallas); Cheng
  Lin (University of Hong Kong); Xiaohu Guo (University of Texas at Dallas)
 ; Shiqing Xin (Shandong University); Wenping Wang (Texas A&M University); 
 and Jing Hua and Zichun Zhong (Wayne State University)\n------------------
 ---\nControllable Shape Modeling with Neural Generalized Cylinder\n\nNeura
 l shape representation, such as neural signed distance field (NSDF), becom
 es more and more popular in shape modeling as its ability to deal with com
 plex topology and arbitrary resolution. Due to the implicit manner to use 
 features for shape representation, manipulating the shapes faces inherent.
 ..\n\n\nXiangyu Zhu (Chinese University of Hong Kong, Shenzhen); Zhiqin Ch
 en (Adobe Research); Ruizhen Hu (Shenzhen University (SZU)); and Xiaoguang
  Han (Chinese University of Hong Kong, Shenzhen)\n---------------------\nD
 irect Manipulation of Procedural Implicit Surfaces\n\nProcedural implicit 
 surfaces are a popular representation for shape modeling. They provide a s
 imple framework for complex geometric operations such as Booleans, blendin
 g and deformations. However, their editability remains a challenging task:
  as the definition of the shape is purely implicit, direct...\n\n\nMarzia 
 Riso (Sapienza University of Rome, Adobe); Élie Michel, Axel Paris, Valent
 in Deschaintre, and Mathieu Gaillard (Adobe); and Fabio Pellacini (Univers
 ity of Modena and Reggio Emilia)\n---------------------\nNeural Laplacian 
 Operator for 3D Point Clouds\n\nThe Laplacian operator holds a crucial rol
 e in 3D geometry processing, yet it is still challenging to define it on p
 oint clouds.\nPrevious works mainly focused on constructing a local triang
 ulation around each point to approximate the underlying manifold for defin
 ing the Laplacian operator, which may...\n\n\nBo Pang, Zhongtian Zheng, Yi
 long Li, Guoping Wang, and Peng-Shuai Wang (Peking University)\n----------
 -----------\nSRIF: Semantic Shape Registration Empowered by Diffusion-base
 d Image Morphing and Flow Estimation\n\nIn this paper, we propose \textbf{
 SRIF}, a novel \textbf{S}emantic shape \textbf{R}egistration framework bas
 ed on diffusion-based \textbf{I}mage morphing and \textbf{F}low Estimation
 . \nMore concretely, given a pair of extrinsically aligned shapes, we firs
 t render them from multi-views, and then we u...\n\n\nMingze Sun (Tsinghua
  shenzhen international graduate school); Chen Guo and Puhua Jiang (Tsingh
 ua shenzhen international graduate school, Pengcheng Lab); and Shiwei Mao,
  Yurun Chen, and Ruqi Huang (Tsinghua shenzhen international graduate scho
 ol)\n---------------------\nHierarchical Light Sampling with Accurate Sphe
 rical Gaussian Lighting\n\nImportance sampling using a light tree (i.e., a
  hierarchy of light clusters) has been widely used for many-light renderin
 g. This technique samples a light source by stochastically traversing the 
 tree according to the importance of each node. While this importance shoul
 d be close to the illumination ...\n\n\nYusuke Tokuyoshi, Sho Ikeda, Parit
 osh Kulkarni, and Takahiro Harada (Advanced Micro Devices, Inc.)\n--------
 -------------\nBSDF importance sampling using a diffusion model\n\nMany re
 al-world materials feature complex BSDFs (Bidirectional Scattering Distrib
 ution Functions) that require significant storage, making the use of neura
 l networks to represent BSDFs appealing. Previous neural sampling methods,
  primarily using analytical lobe mixtures and normalizing flows, often ...
 \n\n\nZiyang Fu, Yash Belhe, Haolin Lu, Liwen Wu, Bing Xu, and Tzu-Mao Li 
 (University of California San Diego)\n---------------------\nDARTS: Diffus
 ion Approximated Residual Time Sampling for Time-of-flight Rendering in Ho
 mogeneous Scattering Media\n\nTime-of-flight (ToF) devices have greatly pr
 opelled the advancement of various multi-modal perception applications. Ho
 wever, achieving accurate rendering of time-resolved information remains a
  challenge, particularly in scenes involving complex geometries, diverse m
 aterials and participating media. ...\n\n\nQianyue He, Dongyu Du, Haitian 
 Jiang, and Xin Jin (Tsinghua Shenzhen International Graduate School)\n----
 -----------------\nNeural Product Importance Sampling via Warp Composition
 \n\nAchieving high efficiency in modern photorealistic rendering methods h
 inges on using Monte Carlo sampling distributions that closely approximate
  the illumination integral estimated for every pixel. Samples are typicall
 y generated from a set of simple distributions, each targeting a different
  factor ...\n\n\nJoey Litalien (McGill University) and Miloš Hašan, Fujun 
 Luan, Krishna Mullia, and Iliyan Georgiev (Adobe Research)\n--------------
 -------\nManifold Sampling for Differentiable Uncertainty in Radiance Fiel
 ds\n\nRadiance fields are powerful and, hence, popular models for represen
 ting the appearance of complex scenes. Yet, constructing them based on ima
 ge observations gives rise to ambiguities and uncertainties. We propose a 
 versatile approach for learning Gaussian radiance fields with explicit and
  fine-grai...\n\n\nLinjie Lyu (Max Planck Institute for Informatics), Ayus
 h Tewari (MIT CSAIL), Marc Habermann (Max Planck Institute for Informatics
 ), Shunsuke Saito and Michael Zollhöfer (Meta Codec Avatars Lab), and Thom
 as Leimkühler and Christian Theobalt (Max Planck Institute for Informatics
 )\n---------------------\nA Generalized Ray Formulation For Wave-Optical L
 ight Transport\n\nRay optics is the foundation of modern path tracing and 
 sampling algorithms for computer graphics; crucially, it allows high-perfo
 rmance implementations based on ray tracing. However, many applications of
  interest in computer graphics and computational optics demand a more prec
 ise understanding of l...\n\n\nShlomi Steinberg (University of Waterloo); 
 Ravi Ramamoorthi (NVIDIA, University of California San Diego); Benedikt Bi
 tterli and Eugene d'Eon (NVIDIA); Ling-Qi Yan (University of California Sa
 nta Barbara); and Matt Pharr (NVIDIA)\n---------------------\nStill-Moving
 : Customized Video Generation without Customized Video Data\n\nCustomizing
  text-to-image (T2I) models has seen tremendous progress recently, particu
 larly in areas such as personalization, stylization, and conditional gener
 ation. However, expanding this progress to video generation is still in it
 s infancy, primarily due to the lack of customized video data. \nIn ...\n\
 n\nHila Chefer (Google Research, Tel Aviv University); Shiran Zada, Roni P
 aiss, Ariel Ephrat, Omer Tov, and Michael Rubinstein (Google Research); Li
 or Wolf (Tel Aviv University); Tali Dekel (Google Research, Weizmann Insti
 tute of Science); Tomer Michaeli (Google Research, Technion – Israel Insti
 tute of Technology); and Inbar Mosseri (Google Research)\n----------------
 -----\nFashion-VDM: Video Diffusion Model for Virtual Try-On\n\nWe present
  Fashion-VDM, a video diffusion model (VDM) for generating virtual try-on 
 videos. Given an input garment image and person video, our method aims to 
 generate a high-quality try-on video of the person wearing the given garme
 nt, while preserving the person's identity and motion. Image-based v...\n\
 n\nJohanna Karras (Google Research, University of Washington); Yingwei Li 
 and Nan Liu (Google Research); Luyang Zhu (Google Research, University of 
 Washington); Innfarn Yoo, Andreas Lugmayr, and Chris Lee (Google Research)
 ; and Ira Kemelmacher-Shlizerman (Google Research, University of Washingto
 n)\n---------------------\nLumiere: A Space-Time Diffusion Model for Video
  Generation\n\nWe introduce Lumiere -- a text-to-video diffusion model des
 igned for synthesizing videos that portray realistic, diverse and coherent
  motion -- a pivotal challenge in video synthesis. To this end, we introdu
 ce a Space-Time U-Net architecture that generates the entire temporal dura
 tion of the video a...\n\n\nOmer Bar-Tal (Google Research, Weizmann Instit
 ute of Science); Hila Chefer (Google Research, Tel Aviv University); Omer 
 Tov, Charles Herrmann, Roni Paiss, Shiran Zada, Ariel Ephrat, Junhwa Hur, 
 Guanghui Liu, Amit Raj, Yuanzhen Li, and Michael Rubinstein (Google Resear
 ch); Tomer Michaeli (Google Research, Technion – Israel Institute of Techn
 ology); Oliver Wang and Deqing Sun (Google Research); Tali Dekel (Google R
 esearch, Weizmann Institute of Science); and Inbar Mosseri (Google Researc
 h)\n---------------------\nI2VEdit: First-Frame-Guided Video Editing via I
 mage-to-Video Diffusion Models\n\nThe remarkable generative capabilities o
 f diffusion models have motivated extensive research in both image and vid
 eo editing. Compared to video editing which faces additional challenges in
  the time dimension, image editing has witnessed the development of more d
 iverse, high-quality approaches and mo...\n\n\nWenqi Ouyang (S-Lab for Adv
 anced Intelligence, Nanyang Technological University Singapore); Yi Dong (
 Nanyang Technological University (NTU)); Lei Yang and Jianlou Si (SenseTim
 e); and Xingang Pan (S-Lab for Advanced Intelligence, Nanyang Technologica
 l University Singapore)\n---------------------\nVidPanos: Generative Panor
 amic Videos from Casual Panning Videos\n\nStitching frames of a panning vi
 deo into a panoramic photograph is a well-understood problem for stationar
 y scenes. When objects are moving, however, a still panorama is not enough
  to capture the scene. \nWe present a method for synthesizing a panoramic 
 video from a casually-captured panning video, a...\n\n\nJingwei Ma (Univer
 sity of Washington); Erika Lu, Roni Paiss, and Shiran Zada (Google Deepmin
 d); Aleksander Holynski (University of California Berkeley, Google Deepmin
 d); Tali Dekel (Weizmann Institute of Science, Google Deepmind); Brian Cur
 less (University of Washington, Google Deepmind); and Michael Rubinstein a
 nd Forrester Cole (Google Deepmind)\n---------------------\nTrailBlazer: T
 rajectory Control for Diffusion-Based Video Generation\n\nLarge text-to-vi
 deo (T2V) models such as Sora have the potential to revolutionize visual e
 ffects and the creation of some types of movies. Current T2V models requir
 e tedious trial-and-error experimentation to achieve desired results, howe
 ver. This motivates the search for methods to directly control...\n\n\nWan
 -Duo Kurt Ma (Victoria University of Wellington), J. P. Lewis (NVIDIA Rese
 arch), and W. Bastiaan Kleijn (Victoria University of Wellington)\n-------
 --------------\nToonCrafter: Generative Cartoon Interpolation\n\nWe introd
 uce ToonCrafter, a novel approach that transcends traditional corresponden
 ce-based cartoon video interpolation, paving the way for generative interp
 olation. Traditional methods, that implicitly assume linear motion and the
  absence of complicated phenomena like dis-occlusion, often struggle ...\n
 \n\nJinbo Xing (Chinese University of Hong Kong); Hanyuan Liu (City Univer
 sity of Hong Kong); Menghan Xia, Yong Zhang, Xintao Wang, and Ying Shan (T
 encent); and Tien-Tsin Wong (Monash University, Chinese University of Hong
  Kong)\n---------------------\nSkeleton-Driven Inbetweening of Bitmap Char
 acter Drawings\n\nOne of the primary reasons for the high cost of traditio
 nal animation is the inbetweening process, where artists manually draw eac
 h intermediate frame necessary for smooth motion. Making this process more
  efficient has been at the core of computer graphics research for years, y
 et the industry has ad...\n\n\nKirill Brodt and Mikhail Bessmeltsev (Unive
 rsity of Montreal)\n---------------------\nSKEL-Betweener: a Neural Motion
  Rig for Interactive Motion Authoring\n\nAuthoring 3D motions is a laborio
 us process that requires manipulating and coordinating many control handle
 s over time. Neural motion representations learned from large motion datas
 ets have recently shown impressive capabilities in many motion completion 
 tasks. However, current methods are not desig...\n\n\nDhruv Agrawal (ETH Z
 ürich, DisneyResearch|Studios) and Jakob Buhmann, Dominik Borer, Robert W.
  Sumner, and Martin Guay (DisneyResearch|Studios)\n---------------------\n
 DrawingSpinUp: 3D Animation from Single Character Drawings\n\nAnimating va
 rious character drawings is an engaging visual content creation task. Give
 n a single character drawing, existing animation methods are limited to fl
 at 2D motions and thus lack 3D effects. An alternative solution is to reco
 nstruct a 3D model from a character drawing as a proxy and then re...\n\n\
 nJie ZHOU (City University of Hong Kong), Chufeng XIAO (Hong Kong Universi
 ty of Science and Technology), Miu-Ling LAM (City University of Hong Kong)
 , and Hongbo FU (Hong Kong University of Science and Technology)\n--------
 -------------\nFrom Sim-to-Real: Toward General Event-based Low-light Fram
 e Interpolation with Per-scene Optimization\n\nVideo Frame Interpolation (
 VFI) is important for video enhancement, frame rate up-conversion, and slo
 w-motion generation. The introduction of event cameras, which capture per-
 pixel brightness changes asynchronously, has significantly enhanced VFI ca
 pabilities, particularly for high-speed, nonlinear ...\n\n\nZiran Zhang (Z
 hejiang University, Shanghai Artificial Intelligence Laboratory); Yongrui 
 Ma (Chinese University of Hong Kong, Shanghai Artificial Intelligence Labo
 ratory); Yueting Chen (Zhejiang University); Feng Zhang (Shanghai Artifici
 al Intelligence Laboratory); Jinwei Gu and Tianfan Xue (Chinese University
  of Hong Kong); and Shi Guo (Shanghai Artificial Intelligence Laboratory)\
 n---------------------\nGFFE: G-buffer Free Frame Extrapolation for Low-la
 tency Real-time Rendering\n\nReal-time rendering has been embracing ever-d
 emanding effects, such as ray tracing. However, rendering such effects in 
 high resolution and high frame rate remains challenging. Frame extrapolati
 on methods, which do not introduce additional latency as opposed to frame 
 interpolation methods such as DLS...\n\n\nSongyin Wu (University of Califo
 rnia Santa Barbara); Deepak Vembar, Anton Sochenov, and Selvakumar Panneer
  (Intel Corporation); Sungye Kim (Intel (now AMD)); Anton Kaplanyan (Intel
  Corporation); and Ling-Qi Yan (University of California Santa Barbara)\n-
 --------------------\nGauWN: Gaussian-smoothed Winding Number and its Deri
 vatives\n\nFor a fixed polygon, one can easily determine whether a point i
 s inside or\noutside it using the winding number. However, deforming a giv
 en polygon\nbased on a set of points with expected inside/outside labeling
  is much more\ndifficult. It asks the winding number to be differentiable 
 with respect to lo...\n\n\nHaoran Sun (State Key Laboratory of CAD&CG, Zhe
 jiang University); Jingkai Wang (State Key Laboratory of CAD&CG, Zhejiang 
 University; Shanghai Jiao Tong University); and Hujun Bao and Jin Huang (S
 tate Key Laboratory of CAD&CG, Zhejiang University)\n---------------------
 \nHodge decomposition of vector fields in Cartesian grids\n\nWhile explici
 t representations of shapes such as triangular and tetrahedral meshes are 
 often used for boundary surfaces and 3D volumes bounded by closed surfaces
 , implicit representations of planar regions and volumetric regions define
 d by level-set functions have also found widespread applications ...\n\n\n
 Zhe Su, Guowei Wei, and Yiying Tong (Michigan State University)\n---------
 ------------\nRobust Symmetry Detection via Riemannian Langevin Dynamics\n
 \nSymmetries are ubiquitous across all kinds of objects, whether in nature
  or in man-made creations. While these symmetries may seem intuitive to th
 e human eye, detecting them with a machine is nontrivial due to the vast s
 earch space. Classical geometry-based methods work by aggregating "votes" 
 for ea...\n\n\nJihyeon Je, Jiayi Liu, Guandao Yang, Boyang Deng, Shengqu C
 ai, and Gordon Wetzstein (Stanford University); Or Litany (Technion); and 
 Leonidas Guibas (Stanford University)\n---------------------\nSharpening a
 nd Sparsifying with Surface Hessians\n\nThe L1 Hessian energy measures the
  norm of the Hessian of a function on a surface (and NOT the squared norm,
  as is common with many geometry applications that employ L2).  Its minimi
 zers tend to be locally linear with a sparse set of curved ridges.  We int
 roduce a fully-intrinsic discretization of t...\n\n\nDylan Rowe (Universit
 y of Southern California); Alec Jacobson (University of Toronto, Adobe Res
 earch); and Oded Stein (University of Southern California)\n--------------
 -------\nQuad mesh mechanisms\n\nThis paper provides computational tools f
 or the modeling and design of quad mesh mechanisms, which are meshes allow
 ing continuous flexions under the assumption of rigid faces and hinges in 
 the edges.  We combine methods and results from different areas, namely di
 fferential geometry of surfaces, rigi...\n\n\nCaigui Jiang (Xi'an Jiaotong
  University); Dmitry Lyakhov (KAUST); Florian Rist (TU Wien, KAUST); Helmu
 t Pottmann (KAUST); and Johannes Wallner (Graz University of Technology)\n
 ---------------------\nPortrait Video Editing Empowered by Multimodal Gene
 rative Priors\n\nWe introduce PortraitGen, a powerful portrait video editi
 ng method that achieves consistent and expressive stylization with multimo
 dal prompts. Traditional portrait video editing methods often struggle wit
 h 3D and temporal consistency, and typically lack in rendering quality and
  efficiency. To addre...\n\n\nXuan Gao, Haiyao Xiao, Chenglai Zhong, Shimi
 n Hu, Yudong Guo, and Juyong Zhang (University of Science and Technology o
 f China)\n---------------------\nHyperGAN-CLIP: A Unified Framework for Do
 main Adaptation, Image Synthesis and Manipulation\n\nGenerative Adversaria
 l Networks (GANs), particularly StyleGAN and its variants, have demonstrat
 ed remarkable capabilities in generating highly realistic images. Despite 
 their success, adapting these models to diverse tasks such as domain adapt
 ation, reference-guided synthesis, and text-guided manipu...\n\n\nAbdul Ba
 sit Anees (Koç University), Ahmet Canberk Baykal (University of Cambridge)
 , Muhammed Burak Kizil (Koç University), Duygu Ceylan (Adobe Research), Er
 kut Erdem (Hacettepe University), and Aykut Erdem (Koç University)\n------
 ---------------\nStableNormal: Reducing Diffusion Variance for Stable and 
 Sharp Normal\n\nThis work addresses the challenge of high-quality surface 
 normal estimation from monocular colored inputs (i.e., images and videos),
  a field which has recently been revolutionized by repurposing diffusion p
 riors. However, previous attempts still struggle with stochastic inference
 , conflicting with t...\n\n\nChongjie Ye and Lingteng Qiu (FNii, The Chine
 se University of Hong Kong, Shenzhen; SSE, The Chinese University of Hong 
 Kong, Shenzhen); Xiaodong Gu and Qi Zuo (Alibaba); Yushuang Wu (FNii, The 
 Chinese University of Hong Kong, Shenzhen; SSE, The Chinese University of 
 Hong Kong, Shenzhen); Zilong Dong and Liefeng Bo (Alibaba); Yuliang Xiu (M
 ax Planck Institute for Intelligent Systems); and Xiaoguang Han (SSE, The 
 Chinese University of Hong Kong, Shenzhen; FNii, The Chinese University of
  Hong Kong, Shenzhen)\n---------------------\nStyleCrafter: Taming Stylize
 d Video Diffusion with Reference-Augmented Adapter Learning\n\nText-to-vid
 eo (T2V) models have shown remarkable capabilities in generating diverse v
 ideos. However, they struggle to produce user-desired artistic videos due 
 to (i) text's inherent clumsiness in expressing specific styles and (ii) t
 he generally degraded style fidelity. To address these challenges, ...\n\n
 \nGongye Liu (Tsinghua University); Menghan Xia, Yong Zhang, and Haoxin Ch
 en (Tencent AI lab); Jinbo Xing (Chinese University of Hong Kong); Yibo Wa
 ng (Tsinghua University); Xintao Wang and Ying Shan (Tencent); and Yujiu Y
 ang (Tsinghua University)\n---------------------\nFast High-Resolution Ima
 ge Synthesis with Latent Adversarial Diffusion Distillation\n\nDiffusion m
 odels are the main driver of progress in image and video synthesis, but su
 ffer from slow inference speed. Distillation methods, like the recently in
 troduced adversarial diffusion distillation (ADD) aim to shift the model f
 rom many-shot to single-step inference, albeit at the cost of expen...\n\n
 \nAxel Sauer, Frederic Boesel, Tim Dockhorn, Andreas Blattmann, Patrick Es
 ser, and Robin Rombach (Black Forest Labs)\n---------------------\nMV2MV: 
 Multi-View Image Translation via View-Consistent Diffusion Models\n\nImage
  translation has various applications in computer graphics and computer vi
 sion, aiming to transfer images from one domain to another. Thanks to the 
 excellent generation capability of diffusion models, recent single-view im
 age translation methods achieve realistic results. However, directly appl.
 ..\n\n\nYoucheng Cai, Runshi Li, and Ligang Liu (University of Science and
  Technology of China)\n---------------------\nFrankenstein: Generating Sem
 antic-Compositional 3D Scenes in One Tri-Plane\n\nWe present Frankenstein,
  a diffusion-based framework that can generate semantic-compositional 3D s
 cenes in a single pass. Unlike existing methods that output a single, unif
 ied 3D shape, Frankenstein simultaneously generates multiple separated sha
 pes, each corresponding to a semantically meaningful p...\n\n\nHan Yan (Sh
 anghai Jiao Tong University); Yang Li (Tencent); Zhennan Wu (University of
  Tokyo); Shenzhou Chen, Weixuan Sun, Taizhang Shang, Weizhe Liu, Tian Chen
 , and Xiaqiang Dai (Tencent); Chao Ma (Shanghai Jiao Tong University); Hon
 gdong Li (Australian National University); and Pan Ji (Tencent)\n---------
 ------------\nDIScene: Object Decoupling and Interaction Modeling for Comp
 lex Scene Generation\n\nThis paper reconsiders how to distill knowledge fr
 om pretrained 2D diffusion models to guide 3D asset generation, in particu
 lar to generate complex 3D scenes: it should accept varied inputs, i.e., t
 exts or images, to allow for flexible expression of requirement; objects i
 n the scene should be style-...\n\n\nXiao-Lei Li (BNRist, Department of Co
 mputer Science and Technology, Tsinghua University); Haodong Li (Hong Kong
  University of Science and Technology, Guangzhou); and Hao-Xiang Chen, Tai
 -Jiang Mu, and Shi-Min Hu (BNRist, Department of Computer Science and Tech
 nology, Tsinghua University)\n---------------------\nTriHuman: A Real-time
  and Controllable Tri-plane Representation for Detailed Human Geometry and
  Appearance Synthesis\n\nCreating controllable, photorealistic, and geomet
 rically detailed digital doubles of real humans solely from video data is 
 a key challenge in Computer Graphics and Vision, especially when real-time
  performance is required. Recent methods attach a neural radiance field (N
 eRF) to an articulated struct...\n\n\nHeming Zhu (Max Planck Institute for
  Informatics, Saarland Informatics Campus); Fangneng Zhan (Max Planck Inst
 itute for Informatics); and Christian Theobalt and Marc Habermann (Max Pla
 nck Institute for Informatics; Saarbrücken Research Center for Visual Comp
 uting, Interaction and AI)\n---------------------\nEgoAvatar: Egocentric V
 iew-Driven and Photorealistic Full-body Avatars\n\nImmersive VR telepresen
 ce ideally means being able to interact and communicate with digital avata
 rs that are indistinguishable from and precisely reflect the behaviour of 
 their real counterparts. The core technical challenge is two fold: Creatin
 g a digital double that faithfully reflects the real hu...\n\n\nJianchun C
 hen and Jian Wang (Max Planck Institute for Informatics; Saarbrücken Resea
 rch Center for Visual Computing, Interaction and AI); Yinda Zhang, Rohit P
 andey, and Thabo Beeler (Google Inc.); and Marc Habermann and Christian Th
 eobalt (Max Planck Institute for Informatics; Saarbrücken Research Center 
 for Visual Computing, Interaction and AI)\n---------------------\nLearn to
  Create Simple LEGO Micro Buildings\n\nThis paper presents the first learn
 ing-based generative pipeline for effectively creating 3D LEGO models. Thi
 s task is very challenging due to the lack of dedicated representations an
 d datasets for learning coherently-connected bricks arrangements, as well 
 as an immense design space that is combinat...\n\n\nJiahao Ge, Mingjun Zho
 u, and Chi-Wing Fu (Chinese University of Hong Kong)\n--------------------
 -\nMagicClay: Sculpting Meshes With Generative Neural Fields\n\nThe recent
  developments in neural fields have brought phenomenal capabilities to the
  field of shape generation, but they lack crucial properties, such as incr
 emental control --- a fundamental requirement for artistic work. Triangula
 r meshes, on the other hand, are the representation of choice for mo...\n\
 n\nAmir Barda (Tel Aviv University), Vladimir Kim (Adobe Research), Noam A
 igerman (Université de Montréal), Amit Haim Bermano (Tel Aviv University),
  and Thibault Groueix (Adobe Research)\n---------------------\nVOODOO XP: 
 Expressive One-Shot Head Reenactment for VR Telepresence\n\nWe introduce V
 OODOO XP: a 3D-aware one-shot head reenactment method that can generate hi
 ghly expressive facial expressions from any input driver video and a singl
 e 2D portrait. Our solution is real-time, view-consistent, and can be inst
 antly used without calibration or fine-tuning. We demonstrate ou...\n\n\nP
 hong Tran (MBZUAI); Egor Zakharov (ETH Zurich); Long-Nhat Ho, Adilbek Karm
 anov, and Ariana Bermudez Venegas (MBZUAI); McLean Goldwhite, Aviral Agarw
 al, and Liwen Hu (Pinscreen); Anh Tran (VinAI Research); and Hao Li (MBZUA
 I, Pinscreen)\n---------------------\nTextToon: Real-Time Text Toonify Hea
 d Avatar from Single Video\n\nWe propose TextToon, a method to generate a 
 drivable toonified avatar. Given a short monocular video sequence and a wr
 itten instruction about the avatar style, our model can generate a high-fi
 delity toonified avatar that can be driven in real-time by another video w
 ith arbitrary identities. Existing...\n\n\nLuchuan Song and Lele Chen (Uni
 veristy of Rochester), Celong Liu (Bytedance), Pinxin Liu (University of R
 ochester), and Chenliang Xu (Univeristy of Rochester)\n-------------------
 --\nPersonaTalk: Bring Attention to Your Persona in Visual Dubbing\n\nFor 
 audio-driven visual dubbing, it remains a considerable challenge to uphold
  and highlight speaker's persona while synthesizing accurate lip synchroni
 zation. Existing methods fall short of capturing speaker's unique speaking
  style or preserving facial details. In this paper, we present PersonaTalk
 ...\n\n\nLonghao Zhang, Shuang Liang, Zhipeng Ge, and Tianshu Hu (Bytedanc
 e)\n---------------------\nTALK-Act: Enhance Textural-Awareness for 2D Spe
 aking Avatar Reenactment with Diffusion Model\n\nRecently, 2D speaking ava
 tars have increasingly participated in everyday scenarios due to the fast 
 development of facial animation techniques. However, most existing works n
 eglect the explicit control of human bodies. In this paper, we propose to 
 drive not only the faces but also the torso and gestu...\n\n\nJiazhi Guan 
 (Tsinghua University); Quanwei Yang (University of Science and Technology 
 of China); Kaisiyuan Wang, Hang Zhou, Shengyi He, Zhiliang Xu, Haocheng Fe
 ng, Errui Ding, and Jingdong Wang (Baidu); Hongtao Xie (University of Scie
 nce and Technology of China); Youjian Zhao (Tsinghua University); and Ziwe
 i Liu (Nanyang Technological University (NTU))\n---------------------\nFol
 low-Your-Emoji: Fine-Controllable and Expressive Freestyle Portrait Animat
 ion\n\nWe present Follow-Your-Emoji, a diffusion-based framework for portr
 ait animation, which animates a reference portrait with target landmark se
 quences. The main challenge of portrait animation is to preserve the ident
 ity of the reference portrait and transfer the target expression to this p
 ortrait whi...\n\n\nYue Ma and Hongyu Liu (Hong Kong University of Science
  and Technology); Hongfa Wang and Heng Pan (Tencent); Yingqing He (Hong Ko
 ng University of Science and Technology); Junkun Yuan, Ailing Zeng, and Ch
 engfei Cai (Tencent); Heung-Yeung Shum (Tsinghua University); Wei Liu (Ten
 cent); and Qifeng Chen (Hong Kong University of Science and Technology)\n-
 --------------------\nFabrig: A Cloth-Simulated Transferable 3D Face Param
 eterization\n\nExisting 3D face parameterization methods are limited to hu
 man faces and/or require a large amount of manual work to prepare face-spe
 cific blendshapes. Unfortunately, many of the automated parameterization m
 ethods do not provide local controls for the different facial regions and 
 methods that allow ...\n\n\nChangAn Zhu and Chris Joslin (Carleton Univers
 ity)\n---------------------\nCurly-Cue: Geometric Methods for Highly Coile
 d Hair\n\nWe present geometric methods for generating shapes that are char
 acteristic of highly coiled hair. Different features become visually relev
 ant when hairs are well-approximated by high-frequency helices instead of 
 a low-frequency curves, so we present algorithms for three such phenomena.
  First, a Four...\n\n\nHaomiao Wu and Alvin Shi (Yale University), A.M. Da
 rke (University of California Santa Cruz), and Theodore Kim (Yale Universi
 ty)\n---------------------\nSPARK: Self-supervised Personalized Real-time 
 Monocular Face Capture\n\nFeedforward monocular face capture methods seek 
 to reconstruct posed faces from a single image of a person. Current state 
 of the art approaches have the ability to regress parametric 3D face model
 s in real-time across a wide range of identities, lighting conditions and 
 poses by leveraging large imag...\n\n\nKelian Baert (Technicolor Group, In
 stitut national de recherche en informatique et en automatique (INRIA) Ren
 nes); Shrisha Bharadwaj (Max Planck Institute for Intelligent Systems); Fa
 bien Castan and Benoit Maujean (Technicolor Group); Marc Christie (Institu
 t national de recherche en informatique et en automatique (INRIA)); Victor
 ia Fernández Abrevaya (Max Planck Institute for Intelligent Systems); and 
 Adnane Boukhayma (Institut national de recherche en informatique et en aut
 omatique (INRIA))\n---------------------\nGroomCap: High-Fidelity Prior-Fr
 ee Hair Capture\n\nDespite recent advances in multi-view hair reconstructi
 on, achieving strand-level precision remains a significant challenge due t
 o inherent limitations in existing capture pipelines. We introduce GroomCa
 p, a novel multi-view hair capture method that reconstructs faithful and h
 igh-fidelity hair geome...\n\n\nYuxiao Zhou (ETH Zürich); Menglei Chai, Da
 oye Wang, Sebastian Winberg, Erroll Wood, and Kripasindhu Sarkar (Google I
 nc.); Markus Gross (ETH Zürich); and Thabo Beeler (Google Inc.)\n---------
 ------------\nTowards Unified 3D Hair Reconstruction from Single-View Port
 raits\n\nSingle-view 3D hair reconstruction is challenging, due to the wid
 e range of shape variations among diverse hairstyles. Current state-of-the
 -art methods are specialized in recovering un-braided 3D hairs and often t
 ake braided styles as their failure cases, because of the inherent difficu
 lty to define...\n\n\nYujian Zheng, Yuda Qiu, and Leyang Jin (Chinese Univ
 ersity of Hong Kong, Shenzhen); Chongyang Ma, Haibin Huang, Di Zhang, and 
 Pengfei Wan (Kuaishou Technology); and Xiaoguang Han (Chinese University o
 f Hong Kong, Shenzhen)\n---------------------\nThe Lips, the Teeth, the ti
 p of the Tongue: LTT Tracking\n\nA mesh-based generative model of the inne
 r-mouth system is presented, which includes teeth and gums for the upper a
 nd lower jaw, the tongue, and their placement inside the human head. The m
 odel is capable of capturing person-specific detail, enabling the creation
  of highly accurate avatars that exce...\n\n\nFeisal Rasras, Stanislav Pid
 horskyi, and Tomas Simon (Reality Labs Research); Hallison Paz (Instituto 
 Nacional de Matemática Pura e Aplicada (IMPA)); and He Wen, Jason Saragih,
  and Javier Romero (Reality Labs Research)\n---------------------\nHairmon
 y: Fairness-aware hairstyle classification\n\nWe present a method for pred
 iction of a person's hairstyle from a single image. Despite growing use ca
 ses in user digitization and enrollment for virtual experiences, available
  methods are limited, particularly in the range of hairstyles they can cap
 ture. Human hair is extremely diverse and lacks an...\n\n\nGivi Meishvili,
  James Clemoes, Charlie Hewitt, Zafiirah Hosenie, Xian Xiao, Martin de La 
 Gorce, Tibor Takacs, Tadas Baltrusaitis, Antonio Criminisi, and Chyna McRa
 e (Microsoft); Nina Jablonski (Pennsylvania State University); and Marta W
 ilczkowiak (Microsoft)\n---------------------\nDifferentiating Variance fo
 r Variance-Aware Inverse Rendering\n\nMonte Carlo methods have been widely
  adopted in physics-based rendering.\n    A key property of a Monte Carlo 
 estimator is its variance, which dictates the convergence rate of the esti
 mator.\n    In this paper, we devise a mathematical formulation for deriva
 tives of rendering variance with respect to ...\n\n\nKai Yan (University o
 f California Irvine, Wētā FX); Vincent Pegoraro, Marc Droske, and Jiří Vor
 ba (Wētā FX); and Shuang Zhao (University of California Irvine)\n---------
 ------------\nDifferentiable Owen Scrambling\n\nQuasi-Monte Carlo integrat
 ion is at the core of rendering. This technique estimates the value of an 
 integral by evaluating the integrand at well-chosen sample locations. Thes
 e sample points are designed to cover the domain as uniformly as possible 
 to achieve better convergence rates than purely rand...\n\n\nBastien Doign
 ies (Université Claude Bernard Lyon, LIRIS); David Coeurjolly, Nicolas Bon
 neel, and Julie Digne (CNRS, LIRIS); and Jean-Claude Iehl and Victor Ostro
 moukhov (Université Claude Bernard Lyon, LIRIS)\n---------------------\nNe
 ural Differential Appearance Equations\n\nWe propose a method to reproduce
  dynamic appearance textures with space-stationary but time-varying visual
  statistics.\nWhile most previous work decomposes dynamic textures into st
 atic appearance and motion, we focus on dynamic appearance that results no
 t from motion but variations of fundamental pro...\n\n\nChen Liu and Tobia
 s Ritschel (University College London (UCL))\n---------------------\nDiffe
 rentiable Photon Mapping using Generalized Path Gradients\n\nPhoton mappin
 g is a fundamental and practical Monte Carlo rendering technique for effic
 iently simulating global illumination effects, especially for caustics and
  specular-diffuse-specular (SDS) paths. In this paper, we present the firs
 t differentiable rendering method for photon mapping. The core of...\n\n\n
 Jiankai Xing and Zengyu Li (Tsinghua University), Fujun Luan (Adobe Resear
 ch), and Kun Xu (Tsinghua University)\n---------------------\nMarkov-Chain
  Monte Carlo Sampling of Visibility Boundaries for Differentiable Renderin
 g\n\nPhysics-based differentiable rendering requires estimating boundary p
 ath integrals emerging from the shift of discontinuities (e.g., visibility
  boundaries). Previously, although the mathematical formulation of boundar
 y path integrals has been established, efficient and robust estimation of 
 these int...\n\n\nPeiyu Xu (University of California Irvine); Sai Bangaru 
 (MIT CSAIL, NVIDIA Research); Tzu-Mao Li (University of California San Die
 go); and Shuang Zhao (University of California Irvine)\n------------------
 ---\nA Simple Approach to Differentiable Rendering of SDFs\n\nWe present a
  simple algorithm for differentiable rendering of surfaces represented by 
 Signed Distance Fields (SDF), which makes it easy to integrate rendering i
 nto gradient-based optimization pipelines. To tackle visibility-related de
 rivatives that make rendering non-differentiable, existing physica...\n\n\
 nZichen Wang and Xi Deng (Cornell University), Ziyi Zhang and Wenzel Jakob
  (EPFL), and Steve Marschner (Cornell University)\n---------------------\n
 MiNNIE: a Mixed Multigrid Method for Real-time Simulation of Nonlinear Nea
 r-Incompressible Elastics\n\nWe propose MiNNIE, a simple yet comprehensive
  framework for real-time simulation of nonlinear near-incompressible elast
 ics. To avoid the common volumetric locking issues at high Poisson's ratio
 s of linear finite element methods (FEM), we build MiNNIE upon a mixed FEM
  framework and further incorporat...\n\n\nLiangwang Ruan (Peking Universit
 y), Bin Wang (Beijing Institute for General Artificial Intelligence), Tian
 tian Liu (Taichi Graphics), and Baoquan Chen (Peking University)\n--------
 -------------\nAnalytic rotation-invariant modelling of anisotropic finite
  elements\n\nA new formulation anisotropic elasticity is presented, which 
 also generalizes to isotropy in low-order invariant-expressible form and w
 ith robust behaviour near singularity-inducing states. As a result, we can
  rewrite, simplify and speedup several existing anisotropic and isotropic 
 distortion energi...\n\n\nHuancheng Lin, Floyd M. Chitalu, and Taku Komura
  (University of Hong Kong)\n---------------------\nTrust-Region Eigenvalue
  Filtering for Projected Newton\n\nWe introduce a novel adaptive eigenvalu
 e filtering strategy to stabilize and accelerate the optimization of Neo-H
 ookean energy and its variants under the Projected Newton framework. For t
 he first time, we show that Newton’s method, Projected Newton with eigenva
 lue clamping and Projected Newton...\n\n\nHonglin Chen (Columbia Universit
 y); Hsueh-Ti Derek Liu (Roblox, University of British Columbia); Alec Jaco
 bson (University of Toronto, Adobe Research); David I.W. Levin (University
  of Toronto, NVIDIA); and Changxi Zheng (Columbia University)\n-----------
 ----------\nAccelerate Neural Subspace-Based Reduced-Order Solver of Defor
 mable Simulation by Lipschitz Optimization\n\nReduced-order simulation is 
 an emerging method for accelerating physical simulations with high DOFs, a
 nd recently developed neural-network-based methods with nonlinear subspace
 s have been proven effective in diverse applications as more concise subsp
 aces can be detected. However, the complexity and ...\n\n\nAoran Lyu (Sout
 h China University of Technology, University of Manchester); Shixian Zhao,
  Chuhua Xian, Zhihao Cen, and Hongmin Cai (South China University of Techn
 ology); and Guoxin Fang (Chinese University of Hong Kong)\n---------------
 ------\nNeural Implicit Reduced Fluid Simulation\n\nHigh-fidelity simulati
 on of fluid dynamics is challenging because of the high dimensional state 
 data needed to capture fine details and the large computational cost assoc
 iated with advancing the system in time. We present neural implicit reduce
 d fluid simulation (NIRFS), a reduced fluid simulation t...\n\n\nYuanyuan 
 Tao (McGill University, Huawei Canada); Ivan Puhachov (Université de Montr
 éal); and Derek Nowrouzezahrai and Paul Kry (McGill University)\n---------
 ------------\nNeural Garment Dynamic Super-Resolution\n\nAchieving efficie
 nt, high-fidelity, high-resolution garment simulation is challenging due t
 o its computational demands. Conversely, low-resolution garment simulation
  is more accessible and ideal for low-budget devices like smartphones. In 
 this paper, we introduce a lightweight, learning-based method...\n\n\nMeng
  Zhang and Jun Li (Nanjing University of Science and Technology)\n--------
 -------------\nParticle Laden Fluid on Flow Maps\n\nWe propose a novel fra
 mework for simulating ink as a particle-laden flow using particle flow map
 s. Our method addresses the limitations of existing flow-map techniques, w
 hich struggle with dissipative forces like viscosity and drag, thereby ext
 ending the application scope from solving the Euler equa...\n\n\nZhiqi Li,
  Duowen Chen, and Candong Lin (Georgia Institute of Technology); Jinyuan L
 iu (Dartmouth College); and Bo Zhu (Georgia Institute of Technology)\n----
 -----------------\nSolid-Fluid Interaction on Particle Flow Maps\n\nWe pro
 pose a novel solid-fluid interaction method for coupling elastic solids wi
 th impulse flow maps. Our key idea is to unify the representation of fluid
  and solid components as particle flow maps with different lengths and dyn
 amics. The solid-fluid coupling is enabled by implementing two novel mec..
 .\n\n\nDuowen Chen and Zhiqi Li (Georgia Institute of Technology); Junwei 
 Zhou (Purdue University, University of Michigan); Fan Feng (Dartmouth Coll
 ege); Tao Du (Tsinghua University, Shanghai Qi Zhi Institute); and Bo Zhu 
 (Georgia Institute of Technology)\n---------------------\nAn Eulerian Vort
 ex Method on Flow Maps\n\nWe present an Eulerian vortex method based on th
 e theory of flow maps to simulate the complex vortical motions of incompre
 ssible fluids. Central to our method is the novel incorporation of the flo
 w-map transport equations for line elements, which, in combination with a 
 bi-directional marching scheme...\n\n\nSinan Wang (Georgia Institute of Te
 chnology, University of Hong Kong); Yitong Deng (Stanford University); Mol
 in Deng (Georgia Institute of Technology); Hong-Xing Yu (Stanford Universi
 ty); Junwei Zhou (Purdue University, University of Michigan); Duowen Chen 
 (Georgia Institute of Technology); Taku Komura (University of Hong Kong); 
 Jiajun Wu (Stanford University); and Bo Zhu (Georgia Institute of Technolo
 gy)\n---------------------\nAn Impulse Ghost Fluid Method for Simulating T
 wo-Phase Flows\n\nThis paper introduces a two-phase interfacial fluid mode
 l based on the impulse variable to capture complex vorticity-interface int
 eractions. Our key idea is to leverage bidirectional flow map theory to en
 hance the transport accuracy of both vorticity and interfaces simultaneous
 ly and address their c...\n\n\nYuchen Sun (Georgia Institute of Technology
 ); Linglai Chen (Harvard University); Weiyuan Zeng (Zhejiang University); 
 Tao Du (Tsinghua University, Shanghai Qi Zhi Institute); Shiying Xiong (Zh
 ejiang University); and Bo Zhu (Georgia Institute of Technology)\n--------
 -------------\nA Unified MPM Framework supporting Phase-field Models and E
 lastic-viscoplastic Phase Transition\n\nRecent years have witnessed the ra
 pid deployment of numerous physicsbased modeling and simulation algorithms
  and techniques for fluids, solids, and their delicate coupling in compute
 r animation. However, it still remains a challenging problem to model the 
 complex elastic-viscoplastic behaviors durin...\n\n\nZaili Tu, Chen Li, Zi
 peng Zhao, Long Liu, Chenhui Wang, and Changbo Wang (East China Normal Uni
 versity  School of Computer Science and Technology) and Hong Qin (Stony Br
 ook University)\n---------------------\nFluid Implicit Particles on Coadjo
 int Orbits\n\nWe propose Coadjoint Orbit FLIP (CO-FLIP), a high order accu
 rate, structure preserving fluid simulation method in the hybrid Eulerian-
 Lagrangian framework. We start with a Hamiltonian formulation of the incom
 pressible Euler Equations, and then, using a local, explicit, and high ord
 er divergence free...\n\n\nMohammad Sina Nabizadeh, Ritoban Roy-Chowdhury,
  Hang Yin, Ravi Ramamoorthi, and Albert Chern (University of California Sa
 n Diego)\n---------------------\nFaçAID: A Transformer Model for Neuro-Sym
 bolic Facade Reconstruction\n\nWe introduce a neuro-symbolic transformer-b
 ased model that converts flat, segmented facade structures into procedural
  definitions using a custom-designed split grammar. To facilitate this, we
  first develop a simple split grammar tailored for architectural facades a
 nd then generate a dataset comprisi...\n\n\nAleksander Plocharski (Warsaw 
 University of Technology, IDEAS NCBR); Jan Swidzinski (IDEAS NCBR); Joanna
  Porter-Sobieraj (Warsaw University of Technology); and Przemyslaw Musials
 ki (New Jersey Institute of Technology, IDEAS NCBR)\n---------------------
 \nLarge Scale Farm Scene Modeling from Remote Sensing Imagery\n\nIn this p
 aper we propose a scalable framework for large-scale farm scene modeling t
 hat utilizes remote sensing data, specifically satellite images. Our appro
 ach begins by accurately extracting and categorizing the distributions of 
 various scene elements from satellite images into four distinct layer...\n
 \n\nZhiqi Xiao and Hao Jiang (Institute of Computing Technology, Chinese A
 cademy of Sciences; University of Chinese Academy of Sciences); Zhigang De
 ng (University of Houston); and Ran Li, Wenwei Han, and Zhaoqi Wang (Insti
 tute of Computing Technology, Chinese Academy of Sciences; University of C
 hinese Academy of Sciences)\n---------------------\nReconstruct translucen
 t thin objects from photos\n\nThe joint reconstruction of shape and appear
 ance for translucent objects from real-world data poses a challenge in com
 puter graphics, especially when dealing with complex layered materials lik
 e leaves or paper. The traditional assumption of diffuse transmittance fal
 ls short, and more accurate Monte-...\n\n\nXi Deng (Cornell University); L
 ifan Wu (NVIDIA Research); Bruce Walter (Cornell University); Ravi Ramamoo
 rthi (University of California San Diego, NVIDIA Research); Eugene d'Eon (
 NVIDIA Research); Steve Marschner (Cornell University, NVIDIA Research); a
 nd Andrea Weidlich (NVIDIA Research)\n---------------------\nDreamUDF: Gen
 erating Unsigned Distance Fields from A Single Image\n\nRecent advances in
  diffusion models and neural implicit surfaces have shown promising progre
 ss in generating 3D models. However, existing generative frameworks are li
 mited to closed surfaces, failing to cope with a wide range of commonly se
 en shapes that have open boundaries. In this work, we presen...\n\n\nYu-Ta
 o Liu and Xuan Gao (Institute of Computing Technology, Chinese Academy of 
 Sciences; University of Chinese Academy of Sciences); Weikai Chen (Tencent
  Games); Jie Yang (Institute of Computing Technology, Chinese Academy of S
 ciences; University of Chinese Academy of Sciences); Xiaoxu Meng and Bo Ya
 ng (Tencent Games); and Lin Gao (Institute of Computing Technology, Chines
 e Academy of Sciences; University of Chinese Academy of Sciences)\n-------
 --------------\nStyle-NeRF2NeRF: 3D Style Transfer from Style-Aligned Mult
 i-View Images\n\nWe propose a simple yet effective pipeline for stylizing 
 a 3D scene, harnessing the power of 2D image diffusion models. Given a NeR
 F model reconstructed from a set of multi-view images, we perform 3D style
  transfer by refining the source NeRF model using stylized images generate
 d by a style-aligned ...\n\n\nHaruo Fujiwara (University of Tokyo) and Yus
 uke Mukuta and Tatsuya Harada (University of Tokyo, RIKEN AIP)\n----------
 -----------\nNU-NeRF: Neural Reconstruction of Nested Transparent Objects 
 with Uncontrolled Capture Environment\n\nThe reconstruction of transparent
  objects is a challenging problem due to the highly noncontinuous and rapi
 dly changing surface color caused by refraction. Existing methods rely on 
 special capture devices, dedicated backgrounds, or ground-truth object mas
 ks to provide more priors and reduce the ambi...\n\n\nJia-Mu Sun (Insititu
 te of Computing Technology Chinese Academy of Sciences, KIRI Innovations);
  Tong Wu (Institute of Computing Technology, Chinese Academy of Sciences; 
 University of Chinese Academy of Sciences); Ling-Qi Yan (University of Cal
 ifornia Santa Barbara); and Lin Gao (Institute of Computing Technology, Ch
 inese Academy of Sciences; University of Chinese Academy of Sciences)\n---
 ------------------\nGaussian Surfel Splatting for Live Human Performance C
 apture\n\nHigh-quality real-time rendering using user-affordable capture r
 igs is an essential property of human performance capture systems for real
 -world applications. However, state-of-the-art performance capture methods
  may not yield satisfactory rendering results under a very sparse (e.g., f
 our) capture s...\n\n\nZheng Dong (State Key Laboratory of CAD&CG, Zhejian
 g University); Ke Xu (City University of Hong Kong); Yaoan Gao, Hujun Bao,
  and Weiwei Xu (State Key Laboratory of CAD&CG, Zhejiang University); and 
 Rynson W.H. Lau (City University of Hong Kong)\n---------------------\nGau
 ssianHeads: End-to-End Learning of Drivable Gaussian Head Avatars from Coa
 rse-to-fine Representations\n\nReal-time rendering of human head avatars i
 s a cornerstone of many computer graphics applications, such as augmented 
 reality, video games, and films, to name a few. Recent approaches address 
 this challenge with computationally efficient geometry primitives in a car
 efully calibrated multi-view setup....\n\n\nKartik Teotia (Max Planck Inst
 itute for Informatics, Saarland Informatics Campus); Hyeongwoo Kim (Imperi
 al College London); Pablo Garrido (Flawless AI); Marc Habermann (Max Planc
 k Institute for Informatics, Saarland Informatics Campus); Mohamed Elghari
 b (Max Planck Institute for Informatics); and Christian Theobalt (Max Plan
 ck Institute for Informatics, Saarland Informatics Campus)\n--------------
 -------\nGGHead: Fast and Generalizable 3D Gaussian Heads\n\nLearning 3D h
 ead priors from large 2D image collections is an important step towards hi
 gh-quality 3D-aware human modeling. \nA core requirement is an efficient a
 rchitecture that scales well to large-scale datasets and large image resol
 utions. \nUnfortunately, existing 3D GANs struggle to scale to gene...\n\n
 \nTobias Kirschstein, Simon Giebenhain, and Jiapeng Tang (Technical Univer
 sity of Munich); Markos Georgopoulos (Independent); and Matthias Nießner (
 Technical University of Munich)\n---------------------\nRobust Dual Gaussi
 an Splatting for Immersive Human-centric Volumetric Videos\n\nVolumetric v
 ideo represents a transformative advancement in visual media, enabling use
 rs to freely navigate immersive virtual experiences and narrowing the gap 
 between digital and real worlds. However, the need for extensive manual in
 tervention to stabilize mesh sequences and the generation of exces...\n\n\
 nYuheng Jiang, Zhehao Shen, Yu Hong, Chengcheng Guo, and Yize Wu (Shanghai
 Tech University); Yingliang Zhang (DGene Inc.); and Jingyi Yu and Lan Xu (
 ShanghaiTech University)\n---------------------\nNPGA: Neural Parametric G
 aussian Avatars\n\nThe creation of high-fidelity, digital versions of huma
 n heads is an important stepping stone in the process of further integrati
 ng virtual components into our everyday lives. Constructing such avatars i
 s a challenging research problem, due to a high demand for photo-realism a
 nd real-time rendering ...\n\n\nSimon Giebenhain and Tobias Kirschstein (T
 echnical University of Munich); Martin Rünz (Synthesia); Lourdes Agapito (
 University College London (UCL), Synthesia); and Matthias Nießner (Technic
 al University of Munich, Synthesia)\n---------------------\nURAvatar: Univ
 ersal Relightable Gaussian Codec Avatars\n\nWe present a new approach to c
 reating photorealistic and relightable head avatars from a phone scan with
  unknown illumination. The reconstructed avatars can be animated and relit
  in real time with the global illumination of diverse environments. Unlike
  existing approaches that estimate parametric re...\n\n\nJunxuan Li, Chen 
 Cao, Gabriel Schwartz, Rawal Khirodkar, Christian Richardt, Tomas Simon, Y
 aser Sheikh, and Shunsuke Saito (Reality Labs Research)\n-----------------
 ----\nSpeed-Aware Audio-Driven Speech Animation using Adaptive Windows\n\n
 We present a novel method that can generate realistic speech animations of
  a 3D face from audio using multiple adaptive windows. In contrast to prev
 ious studies that use a fixed size audio window, our method accepts an ada
 ptive audio window as input, reflecting the audio speaking rate to use con
 sist...\n\n\nSunjin Jung (KAIST, Visual Media Lab); Yeongho Seol (NVIDIA);
  Kwanggyoon Seo and Hyeonho Na (KAIST, Visual Media Lab); Seonghyeon Kim (
 KAIST, Visual Media Lab; Anigma Technologies); and Vanessa Tan and Junyong
  Noh (KAIST, Visual Media Lab)\n---------------------\nSIGGesture: General
 ized Co-Speech Gesture Synthesis via Semantic Injection with Large-Scale P
 re-Training Diffusion Models\n\nThe automated synthesis of high-quality 3D
  gestures from speech holds significant value for virtual humans and gamin
 g. Previous methods primarily focus on synchronizing gestures with speech 
 rhythm, often neglecting semantic gestures. These semantic gestures are sp
 arse and follow a long-tailed distri...\n\n\nQingrong Cheng (Tencent AI La
 b, Tencent TIMI L1 Studio) and Xu Li and Xinghui Fu (Tencent AI Lab)\n----
 -----------------\nWaveBlender: Practical Sound-Source Animation in Blende
 d Domains\n\nSynthesizing plausible sound sources for modern physics-based
  animation is exceptionally challenging due to complex animated phenomena 
 that form rapidly moving, deforming, and vibrating interfaces which produc
 e acoustic waves within the air domain. Not only must the methods synthesi
 ze sounds that ar...\n\n\nKangrui Xue (Stanford University); Jui-Hsien Wan
 g and Timothy Langlois (Adobe Research); and Doug James (Stanford Universi
 ty, NVIDIA)\n---------------------\nDance-to-Music Generation with Encoder
 -based Textual Inversion\n\nThe seamless integration of music with dance m
 ovements is essential for communicating the artistic intent of a dance pie
 ce. This alignment also significantly improves the immersive quality of ga
 ming experiences and animation productions. Although there has been remark
 able advancement in creating hig...\n\n\nSifei Li, Weiming Dong, and Yuxin
  Zhang (MAIS, Institute of Automation, Chinese Academy of Sciences; School
  of Artificial Intelligence, University of Chinese Academy of Sciences); F
 an Tang (University of Chinese Academy of Sciences); Chongyang Ma (Kuaisho
 u Technology); Oliver Deussen (University of Konstanz); Tong-Yee Lee (Nati
 onal Cheng-Kung University); and Changsheng Xu (MAIS, Institute of Automat
 ion, Chinese Academy of Sciences; School of Artificial Intelligence, Unive
 rsity of Chinese Academy of Sciences)\n---------------------\nSketching Wi
 th Your Voice: "Non-Phonorealistic" Rendering of Sounds via Vocal Imitatio
 n\n\nWe present a method for automatically producing human-like vocal imit
 ations of sounds: the equivalent of ``sketching,'' but for auditory rather
  than visual representation. Starting with a simulated model of the human 
 vocal tract, we first try generating vocal imitations by tuning the model'
 s control...\n\n\nMatthew Caren, Kartik Chandra, Joshua Tenenbaum, Jonatha
 n Ragan-Kelley, and Karima Ma (Massachusetts Institute of Technology)\n---
 ------------------\nCorrelation-aware Encoder-Decoder with Adapters for SV
 BRDF Acquisition\n\nCapturing materials from the real world avoids laborio
 us manual material authoring. However, recovering high-fidelity Spatially 
 Varying Bidirectional Reflectance Distribution Function (SVBRDF) maps from
  a few captured images is challenging due to its ill-posed nature. Existin
 g approaches have made e...\n\n\nDi Luo and Hanxiao Sun (Nankai University
 ), Lei Ma (Peking University), Jian Yang (Nankai University), and Beibei W
 ang (Nanjing University)\n---------------------\nNFPLight:  Deep SVBRDF Es
 timation via the Combination of Near and Far Field Point Lighting\n\nRecov
 ering spatial-varying bi-directional reflectance distribution function (SV
 BRDF) from a few hand-held captured images has been a challenging task in 
 computer graphics. Benefiting from the learned priors from data, single-im
 age methods can obtain plausible SVBRDF estimation results. However, the .
 ..\n\n\nLi Wang, Lianghao Zhang, Fangzhou Gao, Yuzhen Kang, and Jiawan Zha
 ng (Tianjin University)\n---------------------\nPolarimetric BSSRDF Acquis
 ition of Dynamic Faces\n\nAcquisition and modeling of polarized light refl
 ection and scattering help reveal the shape, structure, and physical chara
 cteristics of an object, which is increasingly important in computer graph
 ics. However, current polarimetric acquisition systems are limited to stat
 ic and opaque objects. Human f...\n\n\nHyunho Ha and Inseung Hwang (Korea 
 Advanced Institute of Science and Technology (KAIST)), Nestor Monzon (Univ
 ersidad de Zaragoza - I3A), Jaemin Cho and Donggun Kim (Korea Advanced Ins
 titute of Science and Technology (KAIST)), Seung-Hwan Baek (POSTECH), Adol
 fo Muñoz and Diego Gutierrez (Universidad de Zaragoza - I3A), and Min H. K
 im (Korea Advanced Institute of Science and Technology (KAIST))\n---------
 ------------\nControlMat: A Controlled Generative Approach to Material Cap
 ture\n\nMaterial reconstruction from a photograph is a key component of 3D
  content creation democratization. We propose to formulate this ill-posed 
 problem as a controlled synthesis one, leveraging the recent progress in g
 enerative deep networks. We present ControlMat, a method which, given a si
 ngle photogr...\n\n\nGiuseppe Vecchio (Adobe Research, University of Catan
 ia) and Rosalie Martin, Arthur Roullier, Adrien Kaiser, Romain Rouffet, Va
 lentin Deschaintre, and Tamy Boubekeur (Adobe Research)\n-----------------
 ----\nAppearance Modeling of Iridescent Feathers with Diverse Nanostructur
 es\n\nMany animals exhibit structural colors, which are often iridescent, 
 meaning that the perceived colors change with illumination conditions and 
 viewing perspectives. Biological iridescence is usually caused by multilay
 ers or other periodic structures in animal tissues, which selectively refl
 ect light ...\n\n\nYunchen Yu (Cornell University); Andrea Weidlich (NVIDI
 A); Bruce Walter (Cornell University); Eugene d'Eon (NVIDIA); and Steve Ma
 rschner (Cornell University, NVIDIA)\n---------------------\nA Dynamic By-
 example BTF Synthesis Scheme\n\nMeasured Bidirectional Texture Function (B
 TF) can faithfully reproduce a realistic appearance but is costly to acqui
 re and store due to its 6D nature (2D spatial and 4D angular). Therefore, 
 it is practical and necessary for rendering to synthesize BTFs from a smal
 l example patch. While previous meth...\n\n\nZilin Xu (University of Calif
 ornia Santa Barbara), Zahra Montazeri (University of Manchester), Beibei W
 ang (Nanjing University), and Ling-Qi Yan (University of California Santa 
 Barbara)\n---------------------\nOccupancy-Based Dual Contouring\n\nWe int
 roduce a dual contouring method that provides state-of-the-art performance
  for occupancy functions while achieving computation times of a few second
 s. Our method is learning-free and carefully designed to maximize the use 
 of GPU parallelization. The recent surge of implicit neural representati..
 .\n\n\nJisung Hwang and Minhyuk Sung (Korea Advanced Institute of Science 
 and Technology (KAIST))\n---------------------\nGaussian Opacity Fields: E
 fficient Adaptive Surface Reconstruction in Unbounded Scenes\n\nRecently, 
 3D Gaussian Splatting (3DGS) has demonstrated impressive novel view synthe
 sis results, while allowing the rendering of high-resolution images in rea
 l-time. However, leveraging 3D Gaussians for surface reconstruction poses 
 significant challenges due to the explicit and disconnected nature o...\n\
 n\nZehao Yu (University of Tübingen, Tübingen AI Center); Torsten Sattler 
 (Czech Technical University in Prague); and Andreas Geiger (University of 
 Tübingen, Tübingen AI Center)\n---------------------\nA class of new tuned
  primal subdivision schemes with high-quality limit surface in extraordina
 ry regions\n\nWe propose a unified tuning framework for primal subdivision
  schemes that are generalizations of odd-degree uniform B-spline surfaces 
 for unstructured quadrilateral meshes of arbitrary topology. The subdivisi
 on of the resulting tuned primal subdivision (TPS) schemes is performed th
 rough efficient re...\n\n\nXu Wang and Weiyin Ma (City University of Hong 
 Kong)\n---------------------\nSING: Stability-Incorporated Neighborhood Gr
 aph\n\nWe introduce the Stability-Incorporated Neighborhood Graph (SING), 
 a novel density-aware structure designed to capture the intrinsic geometri
 c properties of a point set. We improve upon the spheres-of-influence grap
 h by incorporating additional features to offer more flexibility and contr
 ol in encod...\n\n\nDiana Marin (Technical University of Vienna); Amal Dev
  Parakkat (LTCI, Telecom ParisTech; Institut Polytechnique de Paris); Stef
 an Ohrhallinger and Michael Wimmer (Technical University of Vienna); Steve
  Oudot (Institut national de recherche en informatique et en automatique (
 INRIA)); and Pooran Memari (CNRS, Inria, LIX; École Polytechnique, IP-Pari
 s)\n---------------------\nMulti-level Partition of Unity on Differentiabl
 e Moving Particles\n\nWe introduce a differentiable moving particle repres
 entation based on the\nmulti-level partition of unity (MPU) to represent d
 ynamic implicit geome-\ntries. At the core of our representation are two g
 roups of particles, named\nfeature particles and sample particles, which c
 an move in space and produce\n...\n\n\nJinjin He and Taiyuan Zhang (Dartmo
 uth College); Hiroki Kobayashi and Atsushi Kawamoto (Toyota Central R&D La
 bs., Inc.); Yuqing Zhou (Toyota Research Institute of North America); Tsuy
 oshi Nomura (Toyota Central R&D Labs., Inc.); and Bo Zhu (Georgia Institut
 e of Technology)\n---------------------\nBoosting 3D Object Generation thr
 ough PBR Materials\n\nAutomatic 3D content creation has gained increasing 
 attention recently, due to its potential in various applications such as v
 ideo games, film industry, and AR/VR. \nRecent advancements in diffusion m
 odels and multimodal models have notably improved the quality and efficien
 cy of 3D object generation ...\n\n\nYitong Wang (Fudan University, Shangha
 i Artificial Intelligence Laboratory); Xudong Xu (Shanghai Artificial Inte
 lligence Laboratory); Li Ma (Scanline VFX Studio); Haoran Wang (Shanghai J
 iaotong University); and Bo Dai (Shanghai Artificial Intelligence Laborato
 ry)\n---------------------\nFaceMap: Distortion-Driven Perceptual Facial S
 aliency Maps\n\nHumans are uniquely sensitive to faces. Recognizing fine d
 etail in faces plays an important role in social cognition, identity; and 
 it is key to human interaction. In this work, we present the first quantit
 ative study of the relative importance of face regions to human observers.
  We created a datase...\n\n\nZhongshi Jiang, Kishore Venkateshan, Giljoo N
 am, Meixu Chen, Romain Bachy, Jean-Charles Bazin, and Alexandre Chapiro (M
 eta)\n---------------------\nEnhancing the Aesthetics of 3D Shapes via Ref
 erence-based Editing\n\nWhile there have been previous works that explored
  methods to enhance the aesthetics of images, the automated beautification
  of 3D shapes has been limited to specific shapes such as 3D face models. 
 In this paper, we introduce a framework to automatically enhance the aesth
 etics of general 3D shapes. ...\n\n\nMinchan Chen and Manfred Lau (City Un
 iversity of Hong Kong)\n---------------------\nControlled Spectral Uplifti
 ng for Indirect-Light-Metamerism\n\nSpectral rendering has received increa
 sing attention in recent years. Yet, solutions to define spectral reflecta
 nces are mostly limited to techniques which deterministically uplift exist
 ing RGB inputs. Only recently has uplifting enabled constraining a surface
  appearance under direct illuminants. Ye...\n\n\nMark van de Ruit and Elma
 r Eisemann (Delft University of Technology)\n---------------------\nProced
 ural Material Generation with Reinforcement Learning\n\nModern 3D content 
 creation heavily relies on procedural assets. In particular, procedural ma
 terials are ubiquitous in the industry, but their manipulation remains cha
 llenging. Previous work conditionally generates procedural graphs that mat
 ch a given input image. However, the parameter generation st...\n\n\nBeich
 en Li (MIT CSAIL, Adobe Research); Yiwei Hu, Paul Guerrero, and Milos Hasa
 n (Adobe Research); Liang Shi (MIT CSAIL); Valentin Deschaintre (Adobe Res
 earch); and Wojciech Matusik (MIT CSAIL)\n---------------------\nGPU Corou
 tines for Flexible Splitting and Scheduling of Rendering Tasks\n\nWe intro
 duce coroutines into GPU kernel programming, providing an automated soluti
 on for flexible splitting and scheduling of rendering tasks. This approach
  addresses a prevalent challenge in harnessing the power of modern GPUs fo
 r complex, imbalanced graphics workloads like path tracing. Usually, t...\
 n\n\nShaokun Zheng, Xin Chen, and Zhong Shi (Tsinghua University); Ling-Qi
  Yan (University of California Santa Barbara); and Kun Xu (Tsinghua Univer
 sity)\n---------------------\n360-degree Human Video Generation with 4D Di
 ffusion Transformer\n\nWe present a novel approach for generating 360-degr
 ee high-quality, spatio-temporally coherent human videos from a single ima
 ge. Our framework combines the strengths of diffusion transformers for cap
 turing global correlations across viewpoints and time, and CNNs for accura
 te condition injection. The...\n\n\nRuizhi Shao, Youxin Pang, Zerong Zheng
 , Jingxiang Sun, and Yebin Liu (Tsinghua University)\n--------------------
 -\nSynchronize Dual Hands for Physics-Based Dexterous Guitar Playing\n\nWe
  present a novel approach to synthesize dexterous motions for physically s
 imulated hands in tasks that require coordination between the control of t
 wo hands with high temporal precision. Instead of directly learning a join
 t policy to control two hands, our approach performs bimanual control thro
 ug...\n\n\nPei Xu and Ruocheng Wang (Stanford University)\n---------------
 ------\nWorld-Grounded Human Motion Recovery via Gravity-View Coordinates\
 n\nWe present a novel method for recovering world-grounded human motion fr
 om monocular video. The main challenge lies in the ambiguity of defining t
 he world coordinate system, which varies between sequences. Previous appro
 aches attempt to alleviate this issue by predicting relative motion in an 
 autoreg...\n\n\nZehong Shen, Huaijin Pi, Yan Xia, Zhi Cen, and Sida Peng (
 State Key Laboratory of CAD&CG, Zhejiang University); Zechen Hu (Deep Glin
 t); Hujun Bao (State Key Laboratory of CAD&CG, Zhejiang University); Ruizh
 en Hu (Shenzhen University (SZU)); and Xiaowei Zhou (State Key Laboratory 
 of CAD&CG, Zhejiang University)\n---------------------\nDiffH2O: Diffusion
 -Based Synthesis of Hand-Object Interactions from Textual Descriptions\n\n
 We introduce DiffH2O, a new diffusion-based framework for synthesizing rea
 listic, dexterous hand-object interactions from natural language. Our mode
 l employs a temporal two-stage diffusion process, dividing hand-object mot
 ion generation into grasping and interaction stages to enhance generalizat
 ion ...\n\n\nSammy Christen (ETH, Meta) and Shreyas Hampali, Fadime Sener,
  Edoardo Remelli, Tomas Hodan, Eric Sauser, Shugao Ma, and Bugra Tekin (Me
 ta)\n---------------------\nPuzzleAvatar: Assembling 3D Avatars from Perso
 nal Albums\n\nGenerating personalized 3D avatars is crucial for AR/VR. How
 ever, recent text-to-3D methods that generate avatars for celebrities or f
 ictional characters, struggle with everyday people. Methods for faithful r
 econstruction typically require full-body images in controlled settings. W
 hat if a user coul...\n\n\nYuliang Xiu (Max Planck Institute for Intellige
 nt Systems); Yufei Ye (Carnegie Mellon University); Zhen Liu (Max Planck I
 nstitute for Intelligent Systems; Mila, Université de Montréal); Dimitris 
 Tzionas (University of Amsterdam); and Michael J. Black (Max Planck Instit
 ute for Intelligent Systems)\n---------------------\nMeasuring Human Motio
 n Under Clothing\n\nIn many applications involving clothing, it is essenti
 al to know how clothes move relative to the body hidden beneath. This is i
 mpossible using traditional optical motion capture methods due to visual o
 cclusion due to clothes and other body parts, and very difficult using pre
 vious non-optical method...\n\n\nLuis Bolanos, Pearson Wyder-Hodge, and Xi
 ndong Lin (University of British Columbia) and Dinesh K. Pai (University o
 f British Columbia, Vital Mechanics Research)\n---------------------\nEval
 uating Visual Perception of Object Motion in Dynamic Environments\n\nPreci
 sely understanding how objects move in 3D is essential for broad scenarios
  such as video editing, gaming, driving, and athletics. With screen-displa
 yed computer graphics content, users only perceive limited cues to judge t
 he object motion from the on-screen optical flow. Conventionally, visual .
 ..\n\n\nBudmonde Duinkharjav (New York University); Jenna Kang (Georgia In
 stitute of Technology, New York University); Gavin Stuart Peter Miller and
  Chang Xiao (Adobe Research); and Qi Sun (New York University)\n----------
 -----------\nCasper DPM: Cascaded Perceptual Dynamic Projection Mapping on
 to Hands\n\nWe present a technique for dynamically projecting 3D content o
 nto human hands with short perceived motion-to-photon latency. Computing t
 he pose and shape of human hands accurately and quickly is a challenging t
 ask due to their articulated and deformable nature. We combine a slower 3D
  coarse estimati...\n\n\nYotam Erel and Or Kozlovsky-Mordenfeld (Tel Aviv 
 University), Daisuke Iwai and Kosuke Sato (Osaka University), and Amit H. 
 Bermano (Tel Aviv University)\n---------------------\nIdentifying Behavior
 al Correlates to Visual Discomfort\n\nOutside of self-report surveys, ther
 e are no proven, reliable methods to quantify visual discomfort or visuall
 y induced motion sickness symptoms when using head-mounted displays. While
  valuable tools, self-report surveys suffer from potential biases and low 
 sensitivity due to variability in how resp...\n\n\nDavid Tovar (Reality La
 bs, Meta; Vanderbilt University); James Wilmott and Xiuyun Wu (Reality Lab
 s, Meta); Daniel Martin (Reality Labs Research, Meta; Universidad de Zarag
 oza); and Michael Proulx, Dave Lindberg, Yang Zhao, Olivier Mercier, and P
 hillip Guan (Reality Labs Research, Meta)\n---------------------\nThermOuc
 h: A Wearable Thermo-Haptic Device for Inducing Pain Sensation in Virtual 
 Reality through Thermal Grill Illusion\n\nExisting research showed that un
 pleasant haptic feedback, such as pain, could enhance the user experience 
 and performance in various scenarios (e.g. entertainment and training). Th
 is paper introduces ThermOuch, a wearable thermohaptic device that leverag
 es the thermal grill illusion (TGI) to simulat...\n\n\nHaichen Gao (City U
 niversity of Hong Kong), Shaoyu Cai (National University of Singapore), an
 d Yuhong Wu and Kening Zhu (City University of Hong Kong)\n---------------
 ------\niSeg: Interactive 3D Segmentation via Interactive Attention\n\nWe 
 present iSeg, a new interactive technique for segmenting 3D shapes. Previo
 us works have focused mainly on leveraging pre-trained 2D foundation model
 s for 3D segmentation based on text. However, text may be insufficient for
  accurately describing fine-grained spatial segmentations. Moreover, achie
 v...\n\n\nItai Lang, Fei Xu, and Dale Decatur (University of Chicago); Sud
 arshan Babu (Toyota Technological Institute at Chicago (TTIC)); and Rana H
 anocka (University of Chicago)\n\nRegistration Category: Enhanced Access, 
 Full Access, Full Access Supporter\n\nLanguage Format: English Language
END:VEVENT
END:VCALENDAR