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:20250110T023312Z
LOCATION:Hall B5 (2)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241204T130000
DTEND;TZID=Asia/Tokyo:20241204T141000
UID:siggraphasia_SIGGRAPH Asia 2024_sess116@linklings.com
SUMMARY:It's All About Change: Image Editing
DESCRIPTION:Technical Papers\n\nEach Paper gives a 10 minute presentation.
 \n\nTurboEdit: Text-Based Image Editing Using Few-Step Diffusion Models\n\
 nDiffusion models have opened the path to a wide range of text-based image
  editing frameworks. However, these typically build on the multi-step natu
 re of the diffusion backwards process, and adapting them to distilled, fas
 t-sampling methods has proven surprisingly challenging. Here, we focus on 
 a pop...\n\n\nGilad Deutch (Tel Aviv University); Rinon Gal (Tel Aviv Univ
 ersity, NVIDIA Research); and Daniel Garibi, Or Patashnik, and Daniel Cohe
 n-Or (Tel Aviv University)\n---------------------\nInstantDrag: Improving 
 Interactivity in Drag-based Image Editing\n\nDrag-based image editing has 
 recently gained popularity for its interactivity and precision. However, d
 espite the ability of text-to-image models to generate samples within a se
 cond, drag editing still lags behind due to the challenge of accurately re
 flecting user interaction while maintaining image...\n\n\nJoonghyuk Shin (
 Seoul National University), Daehyeon Choi (POSTECH), and Jaesik Park (Seou
 l National University)\n---------------------\nGenerative Portrait Shadow 
 Removal\n\nWe introduce a high-fidelity portrait shadow removal model that
  can effectively enhance the image of a portrait by predicting its appeara
 nce under disturbing shadows and highlights. Portrait shadow removal is a 
 highly ill-posed problem where multiple plausible solutions can be found b
 ased on a singl...\n\n\nJae Shin Yoon, Zhixin Shu, Mengwei Ren, Xuaner Zha
 ng, Yannick Hold-Geoffroy, Krishna Kumar Singh, and He Zhang (Adobe Inc.)\
 n---------------------\nDiffUHaul: A Training-Free Method for Object Dragg
 ing in Images\n\nText-to-image diffusion models have proven effective for 
 solving many image editing tasks.\n    However, the seemingly straightforw
 ard task of seamlessly relocating objects within a scene remains surprisin
 gly challenging. Existing methods addressing this problem often struggle t
 o function reliably in...\n\n\nOmri Avrahami (Hebrew University of Jerusal
 em), Rinon Gal (Tel Aviv University), Gal Chechik (NVIDIA), Ohad Fried (Th
 e Interdisciplinary Center Herzliya), Dani Lischinski (Hebrew University o
 f Jerusalem), and Arash Vahdat and Weili Nie (NVIDIA)\n-------------------
 --\nContent-aware Tile Generation using Exterior Boundary Inpainting\n\nWe
  present a novel and flexible learning-based method for generating tileabl
 e image sets.  Our method goes beyond simple self-tiling, supporting sets 
 of mutually tileable images that exhibit a high degree of diversity.  To p
 romote diversity we decouple structure from content by foregoing explicit 
 co...\n\n\nSam Sartor and Pieter Peers (College of William & Mary)\n------
 ---------------\nConsolidating Attention Features for Multi-view Image Edi
 ting\n\nLarge-scale text-to-image models enable a wide range of image edit
 ing techniques, using text prompts or even spatial controls. However, appl
 ying these editing methods to multi-view images depicting a single scene l
 eads to 3D-inconsistent results. In this work, we focus on spatial control
 -based geome...\n\n\nOr Patashnik (Tel Aviv University); Rinon Gal (Tel Av
 iv University, NVIDIA Research); Daniel Cohen-Or (Tel Aviv University); an
 d Jun-Yan Zhu and Fernando De La Torre (Carnegie Mellon University)\n\nReg
 istration Category: Full Access, Full Access Supporter\n\nLanguage Format:
  English Language\n\nSession Chair: Dani Lischinski (Hebrew University of 
 Jerusalem, Google)
END:VEVENT
END:VCALENDAR