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 (1)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241204T135800
DTEND;TZID=Asia/Tokyo:20241204T140900
UID:siggraphasia_SIGGRAPH Asia 2024_sess115_papers_1225@linklings.com
SUMMARY:L3DG: Latent 3D Gaussian Diffusion
DESCRIPTION:Technical Papers\n\nBarbara Roessle (Technical University of M
 unich); Norman Müller, Lorenzo Porzi, Samuel Rota Bulò, and Peter Kontschi
 eder (Meta Reality Labs); and Angela Dai and Matthias Nießner (Technical U
 niversity of Munich)\n\nWe propose L3DG, the first approach for generative
  3D modeling of 3D Gaussians through a latent 3D Gaussian diffusion formul
 ation.\nThis enables effective generative 3D modeling, scaling to generati
 on of entire room-scale scenes which can be very efficiently rendered.\nTo
  enable effective synthesis of 3D Gaussians, we propose a latent diffusion
  formulation, operating in a compressed latent space of 3D Gaussians.\nThi
 s compressed latent space is learned by a vector-quantized variational aut
 oencoder (VQ-VAE), for which we employ a sparse convolutional architecture
  to efficiently operate on room-scale scenes. \nThis way, the complexity o
 f the costly generation process via diffusion is substantially reduced, al
 lowing higher detail on object-level generation, as well as scalability to
  large scenes. \nBy leveraging the 3D Gaussian representation, the generat
 ed scenes can be rendered from arbitrary viewpoints in real-time. \nWe dem
 onstrate that our approach significantly improves visual quality over prio
 r work on unconditional object-level radiance field synthesis and showcase
  its applicability to room-scale scene generation.\n\nRegistration Categor
 y: Full Access, Full Access Supporter\n\nLanguage Format: English Language
 \n\nSession Chair: Peng-Shuai Wang (Peking University)
URL:https://asia.siggraph.org/2024/program/?id=papers_1225&sess=sess115
END:VEVENT
END:VCALENDAR