BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Seoul
X-LIC-LOCATION:Asia/Seoul
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:KST
DTSTART:18871231T000000
DTSTART:19881009T020000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20230103T035312Z
LOCATION:Room 324\, Level 3\, West Wing
DTSTART;TZID=Asia/Seoul:20221209T103000
DTEND;TZID=Asia/Seoul:20221209T120000
UID:siggraphasia_SIGGRAPH Asia 2022_sess173_papers_354@linklings.com
SUMMARY:Scalable multi-class sampling via filtered sliced optimal transpor
 t
DESCRIPTION:Technical Communications, Technical Papers\n\nScalable multi-c
 lass sampling via filtered sliced optimal transport\n\nSALAUN, Geogiev, Se
 idel, Singh\n\nWe propose a continuous domain formulation of Wasserstein b
 arycenters\nfor multi-class (-purpose) point set optimization. Our formula
 tion is sys-\ntematically derived to handle hundreds to thousands of class
 es for different\nsampling applications. We develop a practical optimizati
 on scheme that is\nclosely paired with our formulation. To demonstrate the
  generalizability of\nour framework beyond sampling, we formalize the prob
 lem of blue-noise\nerror distribution as a multi-class problem. This helps
  establish a direct\nconnection between sampling, reconstruction and perce
 ptual filtering in\nrendering. The resulting formulation provide error bou
 nds on the perceptual\nerror which, when optimized for, gives screen space
  blue-noise error dis-\ntribution. We demonstrate the effectiveness of our
  framework on different\nsampling applications like stippling, object plac
 ement and rendering.\n\nRegistration Category: FULL ACCESS, ON-DEMAND ACCE
 SS\n\nLanguage: ENGLISH\n\nFormat: IN-PERSON, ON-DEMAND
URL:https://sa2022.siggraph.org/en/full-program/?id=papers_354&sess=sess17
 3
END:VEVENT
END:VCALENDAR