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:20241203T165300
DTEND;TZID=Asia/Tokyo:20241203T170500
UID:siggraphasia_SIGGRAPH Asia 2024_sess110_papers_1057@linklings.com
SUMMARY:Efficient Neural Path Guiding with 4D Modeling
DESCRIPTION:Technical Papers\n\nHonghao Dong, Rui Su, Guoping Wang, and Sh
 eng Li (Peking University)\n\nPrevious local guiding methods used 3D data 
 structures to model spatial radiance variations but struggled with additio
 nal dimensions in the path integral, such as temporal changes in dynamic s
 cenes. Extending these structures to higher dimensions also proves ineffic
 ient due to the curse of dimensionality. In this study, we investigate the
  potential of compact neural representations to model additional scene dim
 ensions efficiently, thereby enhancing the performance of path guiding in 
 specialized rendering applications, such as distributed effects including 
 motion blur. We present an approach that models a higher dimensional spati
 o-temporal distribution through neural feature decomposition. Additionally
 , we present a cost-effective approximate with lower-dimensional represent
 ation to model only subspace by progressive training strategy. We also inv
 estigate the benefits of modeling correlations with the additional dimensi
 ons on typical distributed ray tracing scenarios, including the motion blu
 r effect in dynamic scenes, as well as spectral rendering. Experimental re
 sults demonstrate the effectiveness of our method on these applications.\n
 \nRegistration Category: Full Access, Full Access Supporter\n\nLanguage Fo
 rmat: English Language\n\nSession Chair: Michael Wimmer (TU Wien)
URL:https://asia.siggraph.org/2024/program/?id=papers_1057&sess=sess110
END:VEVENT
END:VCALENDAR