BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
X-LIC-LOCATION:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260417T190051Z
LOCATION:West Building\, Rooms 118-120
DTSTART;TZID=America/Los_Angeles:20250812T140000
DTEND;TZID=America/Los_Angeles:20250812T153000
UID:siggraph_SIGGRAPH 2025_sess149@linklings.com
SUMMARY:Do it With Style & Fashion
DESCRIPTION:FashionComposer: Compositional Fashion Image Generation\n\nFas
 hionComposer is a flexible model for compositional fashion image generatio
 n, with a universal framework that handles diverse input modalities such a
 s text, human models, and garment images. It personalizes appearance, pose
 , and human figure, using subject-binding attention to integrate reference
  ...\n\n\nSihui Ji, Yiyang Wang, and Xi Chen (The University of Hong Kong)
 ; Xiaogang Xu (The Chinese University of Hong Kong); Hao Luo (DAMO Academy
 , Alibaba Group); and Hengshuang Zhao (The University of Hong Kong)\n-----
 ----------------\nDo it With Style & Fashion - Interactive Discussion\n\nA
 fter the summary presentations, attendees will participate in an interacti
 ve discussion. Outside the room will be a series of poster boards for auth
 ors to gather around with the audience. Authors are invited to bring any m
 aterial related to their paper that could instigate further conversation s
 uch...\n\n---------------------\n3D Stylization via Large Reconstruction M
 odel\n\nGiven a 3D object representing the source content and a reference 
 style image, our method performs 3D stylization with a large pre-trained r
 econstruction model. This is achieved in a zero-shot manner, with no train
 ing or test time optimization required, while delivering superior visual f
 idelity and ...\n\n\nIpek Oztas (Bilkent University), Duygu Ceylan (Adobe 
 Research), and Aysegul Dundar (Bilkent University)\n---------------------\
 nStyle Customization of Text-to-Vector Generation with Image Diffusion Pri
 ors\n\nWe propose a novel text-to-vector pipeline with style customization
  that disentangles content and style in SVG generation. Our method represe
 nts the first feed-forward text-to-vector diffusion model capable of gener
 ating SVGs in custom styles.\n\n\nPeiying Zhang (City University of Hong K
 ong), Nanxuan Zhao (Adobe Research), and Jing Liao (City University of Hon
 g Kong)\n---------------------\nStable-Makeup: When Real-World Makeup Tran
 sfer Meets Diffusion Model\n\nStable-Makeup is a diffusion-based makeup tr
 ansfer method. It leverages a Detail-Preserving makeup encoder, and conten
 t-structure control modules to preserve facial content and structure durin
 g transfer. Extensive experiments show that Stable-Makeup outperforms exis
 ting methods, offering robust, gen...\n\n\nYuxuan Zhang (Shanghai Jiao Ton
 g University), Yirui Yuan (Shanghai Tech University), Yiren Song (National
  University of Singapore), and Jiaming Liu (Tiamat AI)\n------------------
 ---\nCobra: Efficient Line Art COlorization with BRoAder References\n\nCob
 ra is a novel efficient long-context fine-grained ID preservation framewor
 k for line art colorization, achieving high precision, efficiency, and fle
 xible usability for comic colorization. By effectively integrating extensi
 ve contextual references, it transforms black-and-white line art into vibr
 a...\n\n\nJunhao Zhuang (Tsinghua University); Lingen Li, Xuan Ju, and Zha
 oyang Zhang (Chinese University of Hong Kong); Chun Yuan (Tsinghua Univers
 ity); and Ying Shan (Tencent)\n---------------------\nReStyle3D: Scene-Lev
 el Appearance Transfer with Semantic Correspondences\n\nRedesign spaces ef
 fortlessly-ReStyle3D transforms indoor scenes by transferring object-speci
 fic styles from a single reference image, preserving 3D coherence. Combini
 ng semantic-aware diffusion and depth guidance, it enables photo-realistic
  virtual staging—faithfully redecorating furniture, te...\n\n\nLiyuan Zhu 
 and Shengqu Cai (Stanford University), Shengyu Huang (NVIDIA), Gordon Wetz
 stein (Stanford University), Naji Khosravan (Zillow), and Iro Armeni (Stan
 ford University)\n\nInterest Area: Research & Education\n\nRecording: Live
 streamed, Not Livestreamed, Recorded, Not Recorded\n\nRegistration Categor
 y: Full Conference, Virtual Access, Tuesday\n\nSession Chair: Jingyi Yu (S
 hanghaiTech University, University of Delaware)
END:VEVENT
END:VCALENDAR