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:20260114T163652Z
LOCATION:Meeting Room C4.11\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231214T162500
DTEND;TZID=Australia/Melbourne:20231214T163500
UID:siggraphasia_SIGGRAPH Asia 2023_sess133_papers_842@linklings.com
SUMMARY:DeepBasis: Hand-Held Single-Image SVBRDF Capture via Two-Level Bas
 is Material Model
DESCRIPTION:Li Wang, Lianghao Zhang, Fangzhou Gao, and Jiawan Zhang (Tianj
 in University)\n\nRecovering spatial-varying bi-directional reflectance di
 stribution function (SVBRDF) from a single hand-held captured image has be
 en a meaningful but challenging task in computer graphics. Benefiting from
  the learned data priors, some previous methods can utilize the potential 
 material correlations between image pixels to serve for SVBRDF estimation.
  \nTo further reduce the ambiguity from single-image estimation, it is nec
 essary to integrate additional explicit material correlations. Given the f
 lexible expressive ability of basis material assumption, we propose DeepBa
 sis, a deep-learning-based method integrated with this assumption. It join
 tly predicts basis materials and their blending weights. Then the estimate
 d SVBRDF is their linear combination. To facilitate the extraction of data
  priors, we introduce a two-level basis model to keep the sufficient repre
 sentative while using a fixed number of basis materials. Moreover, conside
 ring the absence of ground-truth basis materials and weights during networ
 k training, we propose a variance-consistency loss and adopt a joint predi
 ction strategy, thereby enabling the existing SVBRDF dataset available for
  training. Additionally, due to the hand-held capture setting, the exact l
 ighting directions are unknown. We model the lighting direction estimation
  as a sampling problem and propose an optimization-based algorithm to find
  the optimal estimation. Quantitative evaluation and qualitative analysis 
 demonstrate that DeepBasis can produce a higher quality SVBRDF estimation 
 than previous methods. All source codes will be publicly released.\n\nRegi
 stration Category: Full Access\n\nSession Chair: Anton Kaplanyan (Intel)\n
 \n
URL:https://asia.siggraph.org/2023/full-program?id=papers_842&sess=sess133
END:VEVENT
END:VCALENDAR