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:20250110T023312Z
LOCATION:Hall B5 (2)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241203T171600
DTEND;TZID=Asia/Tokyo:20241203T172800
UID:siggraphasia_SIGGRAPH Asia 2024_sess110_papers_1049@linklings.com
SUMMARY:Dynamic Neural Radiosity with Multi-grid Decomposition
DESCRIPTION:Technical Papers\n\nRui Su, Honghao Dong, Jierui Ren, Haojie J
 in, Yisong Chen, Guoping Wang, and Sheng Li (Peking University)\n\nPrior a
 pproaches to the neural rendering of global illumination typically rely on
  complex network architectures and training strategies to model the global
  effects. This often leads to impractically high overheads for both traini
 ng and inference. The neural radiosity technique marks a significant advan
 cement by injecting the radiometric prior into the training process, allow
 ing for efficient modeling of the global radiance fields using a lightweig
 ht network and grid-based representations. However, this method encounters
  difficulties in modeling dynamic scenes, as the high-dimensional feature 
 space quickly becomes unmanageable as the number of varying scene paramete
 rs grows. In this work, we extend neural radiosity for variable scenes thr
 ough a novel neural decomposition method. To achieve this, we first parame
 terize the animated scene with an explicit vector $\mathbf{v}$, which cond
 itions a high-dimensional radiance field $L_{\theta}$. We then develop a p
 ractical representation for $L_{\theta}$ by decomposing the high-dimension
 al feature grid into 3D grids, 2D feature planes, and lightweight MLPs. Th
 is strategy effectively models the correlation between 3D spatial features
  and dynamic scene variables, while maintaining a practical memory and com
 putational cost. Experimental results show that our method facilitates eff
 icient dynamic global illumination rendering with practical runtime perfor
 mance, outperforming previous state-of-the-art techniques with both reduce
 d training and inference costs.\n\nRegistration Category: Full Access, Ful
 l Access Supporter\n\nLanguage Format: English Language\n\nSession Chair: 
 Michael Wimmer (TU Wien)
URL:https://asia.siggraph.org/2024/program/?id=papers_1049&sess=sess110
END:VEVENT
END:VCALENDAR