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:20240214T070240Z
LOCATION:Darling Harbour Theatre\, Level 2 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231212T093000
DTEND;TZID=Australia/Melbourne:20231212T124500
UID:siggraphasia_SIGGRAPH Asia 2023_sess209_papers_842@linklings.com
SUMMARY:DeepBasis: Hand-Held Single-Image SVBRDF Capture via Two-Level Bas
 is Material Model
DESCRIPTION:Technical Papers\n\nLi Wang, Lianghao Zhang, Fangzhou Gao, and
  Jiawan Zhang (Tianjin University)\n\nRecovering spatial-varying bi-direct
 ional reflectance distribution function (SVBRDF) from a single hand-held c
 aptured image has been a meaningful but challenging task in computer graph
 ics. Benefiting from the learned data priors, some previous methods can ut
 ilize the potential material correlations between image pixels to serve fo
 r SVBRDF estimation. \nTo further reduce the ambiguity from single-image e
 stimation, it is necessary to integrate additional explicit material corre
 lations. Given the flexible expressive ability of basis material assumptio
 n, we propose DeepBasis, a deep-learning-based method integrated with this
  assumption. It jointly predicts basis materials and their blending weight
 s. Then the estimated SVBRDF is their linear combination. To facilitate th
 e extraction of data priors, we introduce a two-level basis model to keep 
 the sufficient representative while using a fixed number of basis material
 s. Moreover, considering the absence of ground-truth basis materials and w
 eights during network training, we propose a variance-consistency loss and
  adopt a joint prediction strategy, thereby enabling the existing SVBRDF d
 ataset available for training. Additionally, due to the hand-held capture 
 setting, the exact lighting directions are unknown. We model the lighting 
 direction estimation as a sampling problem and propose an optimization-bas
 ed algorithm to find the optimal estimation. Quantitative evaluation and q
 ualitative analysis demonstrate that DeepBasis can produce a higher qualit
 y SVBRDF estimation than previous methods. All source codes will be public
 ly released.\n\nRegistration Category: Full Access, Enhanced Access, Trade
  Exhibitor, Experience Hall Exhibitor
URL:https://asia.siggraph.org/2023/full-program?id=papers_842&sess=sess209
END:VEVENT
END:VCALENDAR