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:20240214T070248Z
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:Technical Communications, Technical Papers\n\nLi Wang, Liangha
 o Zhang, Fangzhou Gao, and Jiawan Zhang (Tianjin University)\n\nRecovering
  spatial-varying bi-directional reflectance distribution function (SVBRDF)
  from a single hand-held captured image has been a meaningful but challeng
 ing task in computer graphics. Benefiting from the learned data priors, so
 me previous methods can utilize the potential material correlations betwee
 n image pixels to serve for SVBRDF estimation. \nTo further reduce the amb
 iguity from single-image estimation, it is necessary to integrate addition
 al explicit material correlations. Given the flexible expressive ability o
 f basis material assumption, we propose DeepBasis, a deep-learning-based m
 ethod integrated with this assumption. It jointly predicts basis materials
  and their blending weights. Then the estimated SVBRDF is their linear com
 bination. To facilitate the extraction of data priors, we introduce a two-
 level basis model to keep the sufficient representative while using a fixe
 d number of basis materials. Moreover, considering the absence of ground-t
 ruth basis materials and weights during network training, we propose a var
 iance-consistency loss and adopt a joint prediction strategy, thereby enab
 ling the existing SVBRDF dataset available for training. Additionally, due
  to the hand-held capture setting, the exact lighting directions are unkno
 wn. We model the lighting direction estimation as a sampling problem and p
 ropose an optimization-based algorithm to find the optimal estimation. Qua
 ntitative evaluation and qualitative analysis demonstrate that DeepBasis c
 an produce a higher quality SVBRDF estimation than previous methods. All s
 ource codes will be publicly released.\n\nRegistration Category: Full Acce
 ss\n\nSession Chair: Anton Kaplanyan (Intel)
URL:https://asia.siggraph.org/2023/full-program?id=papers_842&sess=sess133
END:VEVENT
END:VCALENDAR