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 (2)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241206T135800
DTEND;TZID=Asia/Tokyo:20241206T140900
UID:siggraphasia_SIGGRAPH Asia 2024_sess146_papers_240@linklings.com
SUMMARY:A Dynamic By-example BTF Synthesis Scheme
DESCRIPTION:Technical Papers\n\nZilin Xu (University of California Santa B
 arbara), Zahra Montazeri (University of Manchester), Beibei Wang (Nanjing 
 University), and Ling-Qi Yan (University of California Santa Barbara)\n\nM
 easured Bidirectional Texture Function (BTF) can faithfully reproduce a re
 alistic appearance but is costly to acquire and store due to its 6D nature
  (2D spatial and 4D angular). Therefore, it is practical and necessary for
  rendering to synthesize BTFs from a small example patch. While previous m
 ethods managed to produce plausible results, we find that they seldomly ta
 ke into consideration the property of being dynamic, so a BTF must be synt
 hesized before the rendering process, resulting in limited size, costly pr
 e-generation and storage issues. In this paper, we propose a dynamic BTF s
 ynthesis scheme, where a BTF at any position only needs to be synthesized 
 when being queried. Our insight is that, with the recent advances in neura
 l dimension reduction methods, a BTF can be decomposed into disjoint low-d
 imensional components. We can perform dynamic synthesis only on the positi
 onal dimensions, and during rendering, recover the BTF by querying and com
 bining these low-dimensional functions with the help of a lightweight Mult
 ilayer Perceptron (MLP). Consequently, we obtain a fully dynamic 6D BTF sy
 nthesis scheme that does not require any pre-generation, which enables eff
 icient rendering of our infinitely large and non-repetitive BTFs on the fl
 y.  We demonstrate the effectiveness of our method through various types o
 f BTFs taken from UBO2014.\n\nRegistration Category: Full Access, Full Acc
 ess Supporter\n\nLanguage Format: English Language\n\nSession Chair: Valen
 tin Deschaintre (Adobe Research)
URL:https://asia.siggraph.org/2024/program/?id=papers_240&sess=sess146
END:VEVENT
END:VCALENDAR