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:20241203T170500
DTEND;TZID=Asia/Tokyo:20241203T171600
UID:siggraphasia_SIGGRAPH Asia 2024_sess110_papers_883@linklings.com
SUMMARY:NeuSmoke: Efficient Smoke Reconstruction and View Synthesis with N
 eural Transportation Fields
DESCRIPTION:Technical Papers\n\nJiaxiong Qiu (TMCC, College of Computer Sc
 ience, Nankai University; Horizon Robotics); Ruihong Cen (TMCC, College of
  Computer Science, Nankai University); Zhong Li (Apple); Han Yan (Nankai T
 MCC, College of Computer Science, Nankai University); and Ming-Ming Cheng 
 and Bo Ren (TMCC, College of Computer Science, Nankai University)\n\nNovel
  view synthesis of smoke scenes presents a challenging problem. Previous n
 eural approaches have suffered from inadequate quality and inefficient tra
 ining. We introduce NeuSmoke, an efficient framework for dynamic smoke rec
 onstruction using neural transportation fields, enabling high-quality dens
 ity reconstruction and novel-view synthesis from multi-view videos. Our fr
 amework consists of two stages. In the first stage, we design a novel neur
 al fluid field representation, integrating the transport equation with neu
 ral transportation fields. This includes adaptive embedding of multiple ti
 me stamps to enhance the spatial-temporal consistency of the reconstructed
  smoke. In the second stage, we combine novel-view color and depth informa
 tion, employing convolutional neural networks (CNNs) to refine the smoke r
 econstruction.  Our model achieves over 10 times faster than previous phys
 ics informed approaches. Extensive experiments demonstrate that our method
  surpasses existing techniques in novel view synthesis and volume density 
 estimation in real-world and synthetic datasets.\n\nRegistration Category:
  Full Access, Full Access Supporter\n\nLanguage Format: English Language\n
 \nSession Chair: Michael Wimmer (TU Wien)
URL:https://asia.siggraph.org/2024/program/?id=papers_883&sess=sess110
END:VEVENT
END:VCALENDAR