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:20250110T023313Z
LOCATION:Hall B5 (1)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241206T110800
DTEND;TZID=Asia/Tokyo:20241206T111900
UID:siggraphasia_SIGGRAPH Asia 2024_sess142_papers_220@linklings.com
SUMMARY:Reconstruct translucent thin objects from photos
DESCRIPTION:Technical Papers\n\nXi Deng (Cornell University); Lifan Wu (NV
 IDIA Research); Bruce Walter (Cornell University); Ravi Ramamoorthi (Unive
 rsity of California San Diego, NVIDIA Research); Eugene d'Eon (NVIDIA Rese
 arch); Steve Marschner (Cornell University, NVIDIA Research); and Andrea W
 eidlich (NVIDIA Research)\n\nThe joint reconstruction of shape and appeara
 nce for translucent objects from real-world data poses a challenge in comp
 uter graphics, especially when dealing with complex layered materials like
  leaves or paper. The traditional assumption of diffuse transmittance fall
 s short, and more accurate Monte-Carlo-based models are often needed to re
 produce their appearance. To accurately capture the translucent appearance
 , an acquisition system needs to be carefully designed. Additionally, ther
 e are three challenges for inverse rendering: First, a large number of unk
 nown parameters make the optimization problem difficult. Second, the Monte
  Carlo (MC) renderer introduces noise, which the optimization is sensitive
  to, \nespecially when dealing with complex material models such as rough 
 dielectric surfaces and highly scattering participating media. Last, MC es
 timators using long light paths (more than 32 bounces in our case) create 
 a large computation graph in memory,  making the gradient back-propagation
  costly.\nTo address those challenges, we present a cheap and fast acquisi
 tion pipeline that can capture spatially-varying reflectance and transmiss
 ion at the same time, using a two-phase optimization. We first initialize 
 the geometry with the traditional vision method and then fit a simple and 
 fast appearance model. Thereafter, we use the estimated parameters to init
 ialize a second optimization using a more expensive volumetric model, whic
 h converges faster and more reliably from this favorable starting position
 .  We also introduce a way to analyze each parameter's sensitivity to the 
 noise in the measurements, which can be used in optimally selecting useful
  measurements for optimization. Furthermore, instead of iterating on the c
 amera system, we also introduce an optimal weighted $\ell_2$ loss as an al
 ternative for selecting useful pixels from existing measurements.\n\nRegis
 tration Category: Full Access, Full Access Supporter\n\nLanguage Format: E
 nglish Language\n\nSession Chair: Maria Larsson (University of Tokyo)
URL:https://asia.siggraph.org/2024/program/?id=papers_220&sess=sess142
END:VEVENT
END:VCALENDAR