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:20260114T163654Z
LOCATION:Meeting Room C4.11\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231214T153000
DTEND;TZID=Australia/Melbourne:20231214T154500
UID:siggraphasia_SIGGRAPH Asia 2023_sess123_papers_317@linklings.com
SUMMARY:Online Scene CAD Recomposition via Autonomous Scanning
DESCRIPTION:Changhao Li and Junfu Guo (University of Science and Technolog
 y of China), Ruizhen Hu (Shenzhen University), and Ligang Liu (University 
 of Science and Technology of China)\n\nAutonomous surface reconstruction o
 f 3D scenes has been intensely studied in recent years, however, it is sti
 ll difficult to accurately reconstruct all the surface details of complex 
 scenes with complicated object relations and severe occlusions, which make
 s the reconstruction results not suitable for direct use in applications s
 uch as gaming and virtual reality. Therefore, instead of reconstructing th
 e detailed surfaces, we aim to recompose the scene with CAD models retriev
 ed from a given dataset to faithfully reflect the object geometry and arra
 ngement in the given scene. Moreover, unlike most of the previous works on
  scene CAD recomposition requiring an offline reconstructed scene or captu
 re video as input, which leads to significant data redundancy, we proposed
  a novel online scene CAD recomposition method with autonomous scanning, w
 hich efficiently recomposes the scene with the guidance of automatically o
 ptimized Next-Best-View (NBV) in a single online scanning pass. Based on t
 he key observation that spatial relation in the scene can not only constra
 in the object pose and layout optimization but also guide the NBV generati
 on, our system consists of two key modules: relation-guided CAD recomposit
 ion module that uses relation-constrained global optimization to get accur
 ate object pose and layout estimation, and relation-aware NBV generation m
 odule that makes the exploration during the autonomous scanning tailored f
 or our composition task. Extensive experiments have been conducted to show
  the superiority of our method over previous methods in scanning efficienc
 y and retrieval accuracy as well as the importance of each key component o
 f our method.\n\nRegistration Category: Full Access\n\nSession Chair: Sai-
 Kit Yeung (Hong Kong University of Science and Technology)\n\n
URL:https://asia.siggraph.org/2023/full-program?id=papers_317&sess=sess123
END:VEVENT
END:VCALENDAR