BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Australia/Melbourne
X-LIC-LOCATION:Australia/Melbourne
BEGIN:DAYLIGHT
TZOFFSETFROM:+1000
TZOFFSETTO:+1100
TZNAME:AEDT
DTSTART:19721003T020000
RRULE:FREQ=YEARLY;BYMONTH=4;BYDAY=1SU
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:19721003T020000
TZOFFSETFROM:+1100
TZOFFSETTO:+1000
TZNAME:AEST
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20240214T070247Z
LOCATION:Meeting Room C4.11\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231214T102000
DTEND;TZID=Australia/Melbourne:20231214T103000
UID:siggraphasia_SIGGRAPH Asia 2023_sess129_papers_820@linklings.com
SUMMARY:Transparent Object Reconstruction via Implicit Differentiable Refr
 action Rendering
DESCRIPTION:Technical Papers\n\nFangzhou Gao, Lianghao Zhang, Li Wang, Jia
 min Cheng, and Jiawan Zhang (Tianjin University)\n\nReconstructing the geo
 metry of transparent objects has been a long-standing challenge. Existing 
 methods rely on complex setups, such as manual annotation or darkroom cond
 itions, to obtain object silhouettes and usually require controlled enviro
 nments with designed patterns to infer ray-background correspondence. Howe
 ver, these intricate arrangements limit the practical application for comm
 on users. In this paper, we significantly simplify the setups and present 
 a novel method that reconstructs transparent objects in unknown natural sc
 enes without manual assistance. Our method incorporates two key technologi
 es. Firstly, we introduce a volume rendering-based method that estimates o
 bject silhouettes by projecting the 3D neural field onto 2D images. This a
 utomated process yields highly accurate multi-view object silhouettes from
  images captured in natural scenes. Secondly, we propose transparent objec
 t optimization through differentiable refraction rendering with the neural
  SDF field, enabling us to optimize the refraction ray based on color rath
 er than explicit ray-background correspondence. Additionally, our optimiza
 tion includes a ray sampling method to supervise the object silhouette at 
 a low computational cost. Extensive experiments and comparisons demonstrat
 e that our method produces high-quality results while offering much more c
 onvenient setups.\n\nRegistration Category: Full Access\n\nSession Chair: 
 Marc Stamminger (Friedrich-Alexander-Universität Erlangen-Nürnberg)
URL:https://asia.siggraph.org/2023/full-program?id=papers_820&sess=sess129
END:VEVENT
END:VCALENDAR