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:20230103T035311Z
LOCATION:Room 324\, Level 3\, West Wing
DTSTART;TZID=Asia/Seoul:20221208T140000
DTEND;TZID=Asia/Seoul:20221208T153000
UID:siggraphasia_SIGGRAPH Asia 2022_sess165_papers_295@linklings.com
SUMMARY:DeepJoin: Learning a Joint Occupancy, Signed Distance, and Normal 
 Field Function for Shape Repair
DESCRIPTION:Technical Papers\n\nDeepJoin: Learning a Joint Occupancy, Sign
 ed Distance, and Normal Field Function for Shape Repair\n\nLamb, Banerjee,
  Banerjee\n\nWe introduce DeepJoin, an automated approach to generate high
 -resolution repairs for fractured shapes using deep neural networks. Exist
 ing approaches to perform automated shape repair operate exclusively on sy
 mmetric objects, require a complete proxy shape, or predict restoration sh
 apes using low-resolution voxels which are too coarse for physical repair.
  We generate a high-resolution restoration shape by inferring a correspond
 ing complete shape and a break surface from an input fractured shape. We p
 resent a novel implicit shape representation for fractured shape repair th
 at combines the occupancy function, signed distance function, and normal f
 ield. We demonstrate repairs using our approach for synthetically fracture
 d objects from ShapeNet, 3D scans from the Google Scanned Objects dataset,
  objects in the style of ancient Greek pottery from the QP Cultural Herita
 ge dataset, and real fractured objects. We outperform six baseline approac
 hes in terms of chamfer distance and normal consistency. Unlike existing a
 pproaches and restorations generated using subtraction, DeepJoin restorati
 ons do not exhibit surface artifacts and join closely to the fractured reg
 ion of the fractured shape. Our code is available at: https://github.com/T
 erascale-All-sensing-Research-Studio/DeepJoin.\n\nRegistration Category: F
 ULL ACCESS, ON-DEMAND ACCESS\n\nLanguage: ENGLISH\n\nFormat: IN-PERSON, ON
 -DEMAND
URL:https://sa2022.siggraph.org/en/full-program/?id=papers_295&sess=sess16
 5
END:VEVENT
END:VCALENDAR