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:20260817T171530Z
LOCATION:Hall B5 (2)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241205T114100
DTEND;TZID=Asia/Tokyo:20241205T115500
UID:siggraphasia_SIGGRAPH Asia 2024_sess128_papers_254@linklings.com
SUMMARY:DirectL: Efficient Radiance Fields Rendering for 3D Light Field Di
 splays
DESCRIPTION:Zongyuan Yang, Baolin Liu, Yingde Song, Lan Yi, Yongping Xiong
 , Zhaohe Zhang, and Xunbo Yu (Beijing University of Posts and Telecommunic
 ations)\n\nAutostereoscopic display technology, despite decades of develop
 ment, has not achieved extensive application, primarily due to the dauntin
 g challenge of three-dimensional (3D) content creation for non-specialists
 . The emergence of Radiance Field as an innovative 3D representation has m
 arkedly revolutionized the domains of 3D reconstruction and generation, si
 mplifying 3D content creation for common users and broadening the applicab
 ility of Light Field Displays (LFDs). However, the combination of these tw
 o technologies remains largely unexplored. The standard paradigm to create
  optimal content for parallax-based light field displays demands rendering
  at least 45 slightly shifted views preferably at high resolution per fram
 e, a substantial hurdle for real-time rendering. We introduce DirectL, a n
 ovel rendering paradigm for Radiance Fields on autostereoscopic displays w
 ith lenticular lens. By thoroughly analyzing the interleaved mapping of sp
 atial rays to screen sub-pixels, we accurately render only the light rays 
 entering the human eye and propose subpixel repurposing to significantly r
 educe the pixel count required for rendering. Tailored for the two predomi
 nant radiance fields—Neural Radiance Fields (NeRFs) and 3D Gaussian Splatt
 ing (3DGS), we propose corresponding optimized rendering pipelines that di
 rectly render the light field images instead of multi-view images, achievi
 ng state-of-the-art rendering speeds on autostereoscopic displays. Extensi
 ve experiments across various autostereoscopic displays and user visual pe
 rception assessments demonstrate that DirectL accelerates rendering by up 
 to 40 times compared to the standard paradigm without sacrificing visual q
 uality. Its rendering process-only modification allows seamless integratio
 n into subsequent radiance field tasks. Finally, we incorporate DirectL in
 to diverse applications, showcasing the stunning visual experiences and th
 e synergy between Light Field Displays and Radiance Fields, which reveals 
 the immense potential for application prospects.\n\nRegistration Category:
  Full Access, Full Access Supporter\n\nLanguage Format: English Language\n
 \nSession Chair: Manolis Savva (Simon Fraser University)\n\n
URL:https://asia.siggraph.org/2024/program/?id=papers_254&sess=sess128
END:VEVENT
END:VCALENDAR