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:20260417T190103Z
LOCATION:West Building\, Rooms 118-120
DTSTART;TZID=America/Los_Angeles:20250814T104500
DTEND;TZID=America/Los_Angeles:20250814T121500
UID:siggraph_SIGGRAPH 2025_sess141@linklings.com
SUMMARY:Very Vivid Videos
DESCRIPTION:Controllable Tracking-Based Video Frame Interpolation\n\nWe pr
 esent a tracking-based video frame interpolation method, optionally guided
  by user inputs. It utilizes sparse point tracks, first estimated using ex
 isting point tracking methods and then optionally refined by the user. Wit
 hout any user input, it already achieves state-of-the-art results, with f.
 ..\n\n\nKarlis Martins Briedis (DisneyResearch|Studios, ETH Zürich); Abdel
 aziz Djelouah and Raphaël Ortiz (DisneyResearch|Studios); Markus Gross (Di
 sneyResearch|Studios, ETH Zürich); and Christopher Schroers (DisneyResearc
 h|Studios)\n---------------------\nVideoAnydoor: High-fidelity Video Objec
 t Insertion with Precise Motion Control\n\nWe propose VideoAnydoor, a zero
 -shot video object insertion framework with high-fidelity detail preservat
 ion and precise motion control, where a pixel warper and a image-video mix
 -training strategy are designed to warp the pixel details according to the
  trajectories. VideoAnydoor demonstrates signif...\n\n\nYuanpeng Tu (The U
 niversity of Hong Kong); Luo Hao (DAMO Academy, Alibaba Group); Chen Xi an
 d Sihui Ji (The University of Hong Kong); Xiang Bai (Huazhong University o
 f Science and Technology); and Zhao Hengshuang (The University of Hong Kon
 g)\n---------------------\nSketch3DVE: Sketch-based 3D-Aware Scene Video E
 diting\n\nWe propose Sketch3DVE, a sketch-based 3D-aware video editing met
 hod to enable detailed local manipulation of videos with significant viewp
 oint changes. Our approach leverages detailed analysis and editing of unde
 rlying 3D scene representations, combined with a diffusion model to synthe
 size realistic...\n\n\nFeng-Lin Liu (Institute of Computing Technology, Ch
 inese Academy of Sciences; University of Chinese Academy of Sciences); Shi
 -Yang Li (Institute of Computing Technology, Chinese Academy of Sciences);
  Yan-Pei Cao (VAST); Hongbo Fu (Hong Kong University of Science and Techno
 logy); and Lin Gao (Institute of Computing Technology, Chinese Academy of 
 Sciences; University of Chinese Academy of Sciences)\n--------------------
 -\nReenact Anything: Semantic Video Motion Transfer Using Motion-Textual I
 nversion\n\nReenact Anything introduces a unified framework for semantic m
 otion transfer, covering applications from full-body and face reenactment 
 to controlling the motion of inanimate objects and the camera. Thereby, mo
 tions are represented using text/image embeddings of an image-to-video dif
 fusion model and...\n\n\nManuel Kansy (ETH Zürich, Disney Research Studios
 ); Jacek Naruniec and Christopher Schroers (Disney Research Studios); Mark
 us Gross (ETH Zürich, Disney Research Studios); and Romann Weber (Disney R
 esearch Studios)\n---------------------\nVery Vivid Videos - Interactive D
 iscussion\n\nAfter the summary presentations, attendees will participate i
 n an interactive discussion. Outside the room will be a series of poster b
 oards for authors to gather around with the audience. Authors are invited 
 to bring any material related to their paper that could instigate further 
 conversation such...\n\n---------------------\nVideoPainter: Any-length Vi
 deo Inpainting and Editing with Plug-and-Play Context Control\n\nVideoPain
 ter introduces a dual-branch framework for video inpainting with a lightwe
 ight context encoder that integrates with pre-trained diffusion transforme
 rs. Its ID resampling strategy maintains identity consistency across any-l
 ength videos, while VPData and VPBench provide the largest segmentati...\n
 \n\nYuxuan Bian (The Chinese University of Hong Kong, Tencent); Zhaoyang Z
 hang (Tencent); Xuan Ju (The Chinese University of Hong Kong); Mingdeng Ca
 o (The University of Tokyo); Liangbin Xie (University of Macau); Ying Shan
  (Tencent); and Qiang Xu (The Chinese University of Hong Kong)\n----------
 -----------\nColorSurge: Bringing Vibrancy and Efficiency to Automatic Vid
 eo Colorization via Dual-Branch Fusion\n\nWe propose ColorSurge, a lightwe
 ight dual-branch network for end-to-end video colorization. It delivers vi
 vid, accurate, and real-time results from grayscale input, and is easily e
 xtensible for high-quality performance at low computational cost.\n\n\nHon
 gbo Zhao, Jiaxing Li, Peiyi Zhang, Peng Xiao, Jianxin Lin, and Yijun Wang 
 (Hunan University)\n\nInterest Area: Research & Education\n\nRecording: Li
 vestreamed, Not Livestreamed, Recorded, Not Recorded\n\nRegistration Categ
 ory: Full Conference, Virtual Access, Thursday\n\nSession Chair: Nanxuan Z
 hao (Adobe Research)
END:VEVENT
END:VCALENDAR