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:20250110T023303Z
LOCATION:G510\, G Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241205T130000
DTEND;TZID=Asia/Tokyo:20241205T135000
UID:siggraphasia_SIGGRAPH Asia 2024_sess288@linklings.com
SUMMARY:Stylization & Evaluation
DESCRIPTION:Technical Communications\n\nThe Technical Communications progr
 am at SIGGRAPH Asia serves as an invaluable platform for presenting cuttin
 g-edge work that may not neatly align with the Technical Papers session. A
 ttendees can expect to explore fresh and thought-provoking ideas, glean pr
 actical insights from real-world production work, and discover innovative 
 applications spanning various disciplines, from graphics and vision to AI 
 and VR.\n\nDuring these sessions, leading experts from academia and indust
 ry will present their latest findings, offering a glimpse into cutting-edg
 e research and development. From geometry and animation to virtual reality
  and machine learning, attendees can expect to explore a diverse array of 
 topics at the intersection of graphics and other fields.\n\nUnder the over
 arching theme of Curious Minds, attendees can delve into discussions surro
 unding innovation, interdisciplinary discovery, and the role of education 
 in shaping the future of technology. Whether you’re a seasoned researcher 
 or a curious enthusiast, the Technical Communications program promises to 
 offer insights that spark curiosity and inspire new perspectives.\n\nA Pra
 ctical Style Transfer Pipeline for 3D Animation: Insights from Production 
 R&D\n\nWe developed a practical style transfer pipeline for creating styli
 zed 3D animation, suitable for real-world production. This paper presents 
 insights from our R&D process, balancing quality, artist control, and work
 load.\n\n\nHideki Todo (Takushoku University); Yuki Koyama, Kunihiro Sakai
 , and Akihiro Komiya (Graphinica, Inc.); and Jun Kato (Arch Inc.)\n-------
 --------------\nAnimateLCM: Computation-Efficient Personalized Style Video
  Generation without Personalized Video Data\n\nComputation-efficient perso
 nalized style video generation without personalized video data, reducing g
 eneration time of similarly sized video diffusion models from 25 seconds t
 o around 1 second while maintaining comparable performance.\n\n\nFu-Yun Wa
 ng (Chinese University of Hong Kong); Zhaoyang Huang (Avolution AI); Weika
 ng Bian, Xiaoyu Shi, and Keqiang Sun (Chinese University of Hong Kong); Gu
 anglu Song (SenseTime); Yu Liu (Shanghai Artificial Intelligence Laborator
 y); and Hongsheng Li (Chinese University of Hong Kong)\n------------------
 ---\nAn Empirical Analysis of GPT-4V’s Performance on Fashion Aesthetic Ev
 aluation\n\nWe examine GPT-4V's zero-shot performance in evaluating outfit
  suitability in images, showing good alignment with human judgments but di
 fficulties in ranking similarly colored outfits.\n\n\nYuki Hirakawa, Takas
 hi Wada, Kazuya Morishita, Ryotaro Shimizu, Takuya Furusawa, Sai Htaung Kh
 am, and Yuki Saito (ZOZO)\n---------------------\nStyleGaussian: Instant 3
 D Style Transfer with Gaussian Splatting\n\nStyleGaussian is a novel 3D st
 yle transfer pipeline that enables instant style transfer while preserving
  real-time rendering and strict multi-view consistency.\n\n\nKunhao Liu (N
 anyang Technological University (NTU)); Fangneng Zhan (Max Planck Institut
 e for Informatics); Muyu Xu (Nanyang Technological University (NTU)); Chri
 stian Theobalt (Max Planck Institute for Informatics); Ling Shao (UCAS-Ter
 minus AI Lab, UCAS); and Shijian Lu (Nanyang Technological University (NTU
 ))\n\nRegistration Category: Full Access, Full Access Supporter\n\nLanguag
 e Format: English Language\n\nSession Chair: Krishna Mullia (Adobe Researc
 h)
END:VEVENT
END:VCALENDAR