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:20260417T190111Z
LOCATION:West Building\, Rooms 301-305
DTSTART;TZID=America/Los_Angeles:20250813T140000
DTEND;TZID=America/Los_Angeles:20250813T153000
UID:siggraph_SIGGRAPH 2025_sess117@linklings.com
SUMMARY:Light & Relight
DESCRIPTION:Practical Inverse Rendering of Textured and Translucent Appear
 ance\n\nThis work addresses recovering textured materials using inverse re
 ndering. Our Laplacian mipmapping improves the reconstruction of high-reso
 lution textures. We also propose a novel gradient computation that enables
  efficiently reconstructing textured, path-traced subsurface scattering. T
 he methods a...\n\n\nPhilippe Weier (Saarland University, Google); Jérémy 
 Riviere, Ruslan Guseinov, and Stephan Garbin (Google); Philipp Slusallek (
 Saarland University, DFKI); Bernd Bickel (Google, ETH Zürich); and Thabo B
 eeler and Delio Vicini (Google)\n---------------------\nLayerPano3D: Layer
 ed 3D Panorama for Hyper-Immersive Scene Generation\n\nLayerPano3D is a no
 vel framework that generates hyper-immersive 3D panoramic scenes from a si
 ngle text prompt. By decomposing panoramas into multiple layers and optimi
 zing them as 3D Gaussians, it enables full 360°×180° exploration with cons
 istent visual quality, unlocking new possibilities for virt...\n\n\nShuai 
 Yang (Shanghai Jiao Tong University, Shanghai Artificial Intelligence Labo
 ratory); Jing Tan (The Chinese University of Hong Kong); Mengchen Zhang (Z
 hejiang University, Shanghai Artificial Intelligence Laboratory); Tong Wu 
 (The Chinese University of Hong Kong); Gordon Wetzstein (Stanford Universi
 ty); Ziwei Liu (Nanyang Technological University); and Dahua Lin (The Chin
 ese University of Hong Kong)\n---------------------\nLight & Relight - Int
 eractive Discussion\n\nAfter the summary presentations, attendees will par
 ticipate in an interactive discussion. Outside the room will be a series o
 f poster boards for authors to gather around with the audience. Authors ar
 e invited to bring any material related to their paper that could instigat
 e further conversation such...\n\n---------------------\nGSHeadRelight: Fa
 st Relightability for 3D Gaussian Head Synthesis\n\nGSHeadRelight enables 
 fast, high-quality relightability for 3D Gaussian head synthesis. A linear
  light model based on learnable radiance transfer is integrated into the n
 ative 3DGS rasterization process and supports colored illumination. Withou
 t requiring expensive light stage data, our method achie...\n\n\nHenglei L
 v (Institute of Computing Technology, Chinese Academy of Sciences; Univers
 ity of Chinese Academy of Sciences); Bailin Deng (Cardiff University); Jia
 nzhu Guo, Xiaoqiang Liu, Pengfei Wan, and Di Zhang (Kuaishou Technology); 
 and Lin Gao (Institute of Computing Technology, Chinese Academy of Science
 s)\n---------------------\nLightLab: Controlling Light Sources in Images w
 ith Diffusion Models\n\nLightLab is a diffusion-based method for parametri
 c control over light sources in an image. Leveraging the linearity of ligh
 t we create a dataset of controled illumniation changes from a small set o
 f real image pairs and synthetic renders, which is used to fine-tune a mod
 el to enable physically plau...\n\n\nNadav Magar (Tel Aviv University, Goo
 gle); Amir Hertz, Eric Tabellion, Yael Pritch, and Alex Rav-Acha (Google);
  Ariel Shamir (Reichman University, Google); and Yedid Hoshen (Hebrew Univ
 ersity of Jerusalem, Google)\n---------------------\nPhysically Controllab
 le Relighting of Photographs\n\nWe present a photograph relighting method 
 that enables explicit control over light sources akin to CG pipelines. We 
 achieve this in a pipeline involving mid-level computer vision, physically
 -based rendering, and neural rendering. We introduce a self-supervised tra
 ining methodology to train our neura...\n\n\nChris Careaga and Yağız Aksoy
  (Simon Fraser University)\n---------------------\nSpatiotemporally Consis
 tent Indoor Lighting Estimation with Diffusion Priors\n\nWe propose a meth
 od for estimating spatiotemporally varying indoor lighting from videos usi
 ng a continuous light field represented as an MLP. By leveraging 2D diffus
 ion priors fine-tuned to predict lighting jointly at multiple locations, o
 ur approach achieves superior performance and zero-shot gener...\n\n\nMuti
 an Tong, Rundi Wu, and Changxi Zheng (Columbia University)\n\nInterest Are
 a: Research & Education\n\nRecording: Livestreamed, Not Livestreamed, Reco
 rded, Not Recorded\n\nRegistration Category: Full Conference, Virtual Acce
 ss, Wednesday\n\nSession Chair: Miloš Hašan (Adobe)
END:VEVENT
END:VCALENDAR