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:20260817T171531Z
LOCATION:Hall B5 (2)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241205T105900
DTEND;TZID=Asia/Tokyo:20241205T111300
UID:siggraphasia_SIGGRAPH Asia 2024_sess128_papers_536@linklings.com
SUMMARY:3D Gaussian Ray Tracing: Fast Tracing of Particle Scenes
DESCRIPTION:Nicolas Moenne-Loccoz (NVIDIA); Ashkan Mirzaei (NVIDIA, Univer
 sity of Toronto); and Or Perel, Riccardo de Lutio, Janick Martinez Esturo,
  Gavriel State, Sanja Fidler, Nicholas Sharp, and Zan Gojcic (NVIDIA)\n\nP
 article-based representations of radiance fields such as 3D Gaussian Splat
 ting have found great success for reconstructing and re-rendering of compl
 ex scenes.\nMost existing methods render particles via rasterization, proj
 ecting them to screen space tiles for processing in a sorted order.\nThis 
 work instead considers ray tracing the particles, building a bounding volu
 me hierarchy and casting a ray for each pixel using high-performance GPU r
 ay tracing hardware.\nTo efficiently handle large numbers of semi-transpar
 ent particles, we describe a specialized rendering algorithm which encapsu
 lates particles with bounding meshes to leverage fast ray-triangle interse
 ctions, and shades batches of intersections in depth-order. \nThe benefits
  of ray tracing are well-known in computer graphics: processing incoherent
  rays for secondary lighting effects such as shadows and reflections, rend
 ering from highly-distorted cameras common in robotics, stochastically sam
 pling rays, and more.\nWith our renderer, this flexibility comes at little
  cost compared to rasterization. Experiments demonstrate the speed and acc
 uracy of our approach, as well as several applications in computer graphic
 s and vision.\nWe further propose related improvements to the basic Gaussi
 an representation, including a simple use of generalized kernel functions 
 which significantly reduces particle hit counts.\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_536&sess=sess128
END:VEVENT
END:VCALENDAR