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:20260805T205450Z
LOCATION:Room 408 B
DTSTART;TZID=America/Los_Angeles:20260722T140000
DTEND;TZID=America/Los_Angeles:20260722T153000
UID:siggraph_SIGGRAPH 2026_sess145@linklings.com
SUMMARY:Film/VFX Production
DESCRIPTION:BodyReLux: Temporally Consistent Full-Body Video Relighting\n\
 nBodyReLux is a video diffusion framework for relighting full-body human p
 erformances with photorealistic, temporally consistent results. Trained on
  video relighting pairs captured on large LED Sphere using digital bi-pack
  technology, it enables precise, dynamic lighting control and generalizes 
 robus...\n\n\nLi MA, Mingming He, Xueming Yu, David George, and Ahmet Tase
 l (Eyeline Labs); Paul Debevec (Eyeline Labs, Netflix); and Julien Philip 
 (Eyeline Labs)\n---------------------\nPixel Cube: Diffusion-based Portrai
 t Video Relighting Through Realistic Lighting Reproduction\n\nPixel Cube i
 s a novel lighting system for realistic lighting emulation and high-speed 
 video relighting data acquisition. With a high-quality lighting-aware data
 set, we train a diffusion model for dynamic portrait relighting. Our model
  produces photorealistic, temporally consistent portrait videos un...\n\n\
 nYufan Zhang (George Mason University), Yu Ji (LightThought LLC), Ayo Ajib
 oye (George Mason University), Rundi Wu (Columbia University), Yu Guo (Geo
 rge Mason University), Changxi Zheng (Columbia University), and Jinwei Ye 
 (George Mason University)\n---------------------\nEasyVFX: Frequency-Drive
 n Decoupling for Resource-Efficient VFX Generation\n\nEasyVFX is a resourc
 e-efficient framework for high-fidelity VFX synthesis. It leverages freque
 ncy-domain decomposition to disentangle spatial detail and motion dynamics
 , reducing learning complexity. A frequency-aware MoE enables efficient tr
 aining, while a test-time adaptation strategy refines unse...\n\n\nYue Ma 
 and Xu Ye (Hong Kong University of Science and Technology); Qinghe Wang (D
 alian University of Technology); Yucheng Wang, Hongyu Liu, and Yinhan Zhan
 g (Hong Kong University of Science and Technology); Xinyu Wang (Tsinghua U
 niversity); Yuanpeng Chen and Shanhui Mo (Independent); Paul Liang and Fan
 gneng Zhan (Massachusetts Institute of Technology (MIT)); and Qifeng Chen 
 (Hong Kong University of Science and Technology)\n---------------------\nL
 ongE2V: Long-Horizon Event-based Video Reconstruction, Prediction, and Fra
 me Interpolation with Video Diffusion Models\n\nThis work introduces long-
 horizon event-based video generation, a challenging task prone to severe t
 emporal inconsistency, and presents LongE2V, a novel approach leveraging v
 ideo diffusion priors to jointly handle video reconstruction, prediction, 
 and frame interpolation, recovering high-quality vid...\n\n\nCheng-De Fan,
  Chun-Wei Tuan Mu, Chen-Wei Chang, Chin-Yang Lin, Kun-Ru Wu, Yu-Chee Tseng
 , and Yu-Lun Liu (National Yang Ming Chiao Tung University)\n-------------
 --------\nAutoregressive Modeling of Film with Applications in Video Monta
 ge\n\nThis work introduces an autoregressive transformer designed for vide
 o montage, transforming raw footage into cinematic sequences. We train it 
 on large corpora of movies to learn film grammar. Instead of generating pi
 xels, we introduce a footage-constrained decoding algorithm to select shot
 s from the...\n\n\nMarcelo Sandoval-Castañeda (TTIC, Adobe); Fabian Caba H
 eilbron (Adobe); Shiry Ginosar (TTIC); Bryan Russell (Adobe); Josef Sivic 
 (Adobe; CTU, Prague); Alexei Efros (University of California Berkeley); an
 d Gregory Shakhnarovich (TTIC)\n---------------------\nFilm/VFX Production
  - Interactive Discussion\n\nAfter the summary presentations, attendees wi
 ll participate in an interactive discussion. Outside the room will be a se
 ries of poster boards for authors to gather around with the audience. Auth
 ors are invited to bring any material related to their paper that could in
 stigate further conversation such...\n\n\nInterest Area: Research & Educat
 ion\n\nKeyword: Artificial Intelligence/Machine Learning, Generative AI, L
 ighting\n\nRegistration Category: Full Conference Supporter, Full Conferen
 ce\n\nSession Chair: Or Patashnik (Tel Aviv University, Snap Research)
END:VEVENT
END:VCALENDAR