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: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:Technical Papers\n\nZongyuan Yang, Baolin Liu, Yingde Song, La
 n Yi, Yongping Xiong, Zhaohe Zhang, and Xunbo Yu (Beijing University of Po
 sts and Telecommunications)\n\nAutostereoscopic display technology, despit
 e decades of development, has not achieved extensive application, primaril
 y due to the daunting challenge of three-dimensional (3D) content creation
  for non-specialists. The emergence of Radiance Field as an innovative 3D 
 representation has markedly revolutionized the domains of 3D reconstructio
 n and generation, simplifying 3D content creation for common users and bro
 adening the applicability of Light Field Displays (LFDs). However, the com
 bination of these two technologies remains largely unexplored. The standar
 d paradigm to create optimal content for parallax-based light field displa
 ys demands rendering at least 45 slightly shifted views preferably at high
  resolution per frame, a substantial hurdle for real-time rendering. We in
 troduce DirectL, a novel rendering paradigm for Radiance Fields on autoste
 reoscopic displays with lenticular lens. By thoroughly analyzing the inter
 leaved mapping of spatial rays to screen sub-pixels, we accurately render 
 only the light rays entering the human eye and propose subpixel repurposin
 g to significantly reduce the pixel count required for rendering. Tailored
  for the two predominant radiance fields—Neural Radiance Fields (NeRFs) an
 d 3D Gaussian Splatting (3DGS), we propose corresponding optimized renderi
 ng pipelines that directly render the light field images instead of multi-
 view images, achieving state-of-the-art rendering speeds on autostereoscop
 ic displays. Extensive experiments across various autostereoscopic display
 s and user visual perception assessments demonstrate that DirectL accelera
 tes rendering by up to 40 times compared to the standard paradigm without 
 sacrificing visual quality. Its rendering process-only modification allows
  seamless integration into subsequent radiance field tasks. Finally, we in
 corporate DirectL into diverse applications, showcasing the stunning visua
 l experiences and the synergy between Light Field Displays and Radiance Fi
 elds, which reveals the immense potential for application prospects.\n\nRe
 gistration Category: Full Access, Full Access Supporter\n\nLanguage Format
 : English Language\n\nSession Chair: Manolis Savva (Simon Fraser Universit
 y)
URL:https://asia.siggraph.org/2024/program/?id=papers_254&sess=sess128
END:VEVENT
END:VCALENDAR