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:20260817T171534Z
LOCATION:Hall B5 (2)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241205T163000
DTEND;TZID=Asia/Tokyo:20241205T174000
UID:siggraphasia_SIGGRAPH Asia 2024_sess137@linklings.com
SUMMARY:Diffuse and Conquer
DESCRIPTION:Each Paper gives a 10 minute presentation.\n\nPortrait Video E
 diting Empowered by Multimodal Generative Priors\n\nWe introduce PortraitG
 en, a powerful portrait video editing method that achieves consistent and 
 expressive stylization with multimodal prompts. Traditional portrait video
  editing methods often struggle with 3D and temporal consistency, and typi
 cally lack in rendering quality and efficiency. To addre...\n\n\nXuan Gao,
  Haiyao Xiao, Chenglai Zhong, Shimin Hu, Yudong Guo, and Juyong Zhang (Uni
 versity of Science and Technology of China)\n---------------------\nHyperG
 AN-CLIP: A Unified Framework for Domain Adaptation, Image Synthesis and Ma
 nipulation\n\nGenerative Adversarial Networks (GANs), particularly StyleGA
 N and its variants, have demonstrated remarkable capabilities in generatin
 g highly realistic images. Despite their success, adapting these models to
  diverse tasks such as domain adaptation, reference-guided synthesis, and 
 text-guided manipu...\n\n\nAbdul Basit Anees (Koç University), Ahmet Canbe
 rk Baykal (University of Cambridge), Muhammed Burak Kizil (Koç University)
 , Duygu Ceylan (Adobe Research), Erkut Erdem (Hacettepe University), and A
 ykut 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 colore
 d inputs (i.e., images and videos), a field which has recently been revolu
 tionized by repurposing diffusion priors. However, previous attempts still
  struggle with stochastic inference, conflicting with t...\n\n\nChongjie Y
 e and Lingteng Qiu (FNii, The Chinese University of Hong Kong, Shenzhen; S
 SE, The Chinese University of Hong Kong, Shenzhen); Xiaodong Gu and Qi Zuo
  (Alibaba); Yushuang Wu (FNii, The Chinese University of Hong Kong, Shenzh
 en; SSE, The Chinese University of Hong Kong, Shenzhen); Zilong Dong and L
 iefeng Bo (Alibaba); Yuliang Xiu (Max Planck Institute for Intelligent Sys
 tems); and Xiaoguang Han (SSE, The Chinese University of Hong Kong, Shenzh
 en; FNii, The Chinese University of Hong Kong, Shenzhen)\n----------------
 -----\nStyleCrafter: Taming Stylized Video Diffusion with Reference-Augmen
 ted Adapter Learning\n\nText-to-video (T2V) models have shown remarkable c
 apabilities in generating diverse videos. However, they struggle to produc
 e user-desired artistic videos due to (i) text's inherent clumsiness in ex
 pressing specific styles and (ii) the generally degraded style fidelity. T
 o address these challenges, ...\n\n\nGongye Liu (Tsinghua University); Men
 ghan Xia, Yong Zhang, and Haoxin Chen (Tencent AI lab); Jinbo Xing (Chines
 e University of Hong Kong); Yibo Wang (Tsinghua University); Xintao Wang a
 nd Ying Shan (Tencent); and Yujiu Yang (Tsinghua University)\n------------
 ---------\nFast High-Resolution Image Synthesis with Latent Adversarial Di
 ffusion Distillation\n\nDiffusion models are the main driver of progress i
 n image and video synthesis, but suffer from slow inference speed. Distill
 ation methods, like the recently introduced adversarial diffusion distilla
 tion (ADD) aim to shift the model from many-shot to single-step inference,
  albeit at the cost of expen...\n\n\nAxel Sauer, Frederic Boesel, Tim Dock
 horn, Andreas Blattmann, Patrick Esser, and Robin Rombach (Black Forest La
 bs)\n---------------------\nMV2MV: Multi-View Image Translation via View-C
 onsistent Diffusion Models\n\nImage translation has various applications i
 n computer graphics and computer vision, aiming to transfer images from on
 e domain to another. Thanks to the excellent generation capability of diff
 usion models, recent single-view image translation methods achieve realist
 ic results. However, directly appl...\n\n\nYoucheng Cai, Runshi Li, and Li
 gang Liu (University of Science and Technology of China)\n\nRegistration C
 ategory: Full Access, Full Access Supporter\n\nLanguage Format: English La
 nguage\n\nSession Chair: Michael Rubinstein (Google)
END:VEVENT
END:VCALENDAR