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:20230103T035307Z
LOCATION:Room 325-AB\, Level 3\, West Wing
DTSTART;TZID=Asia/Seoul:20221206T140000
DTEND;TZID=Asia/Seoul:20221206T153000
UID:siggraphasia_SIGGRAPH Asia 2022_sess156_papers_327@linklings.com
SUMMARY:3QNet: 3D Point Cloud Geometry Quantization Compression Network
DESCRIPTION:Technical Communications, Technical Papers, TOG\n\n3QNet: 3D P
 oint Cloud Geometry Quantization Compression Network\n\nHuang, zhang, Chen
 , Ding, Tai...\n\nSince the development of 3D applications, the point clou
 d, as a spatial description easily acquired by sensors, has been widely us
 ed in multiple areas such as SLAM and 3D reconstruction. Point Cloud Compr
 ession (PCC) has also attracted more attention as a primary step before po
 int cloud transferring and saving, where the geometry compression is an im
 portant component of PCC to compress the points geometrical structures. Ho
 wever, existing non-learning-based geometry compression methods are often 
 limited by manually pre-defined compression rules. Though learning-based c
 ompression methods can significantly improve the algorithm performances by
  learning compression rules from data, they still have some defects. Voxel
 -based compression networks introduce precision errors due to the voxelize
 d operations, while point-based methods may have relatively weak robustnes
 s and are mainly designed for sparse point clouds. In this work, we propos
 e a novel learning-based point cloud compression framework named 3D Point 
 Cloud Geometry Quantiation Compression Network (3QNet), which overcomes th
 e robustness limitation of existing point-based methods and can handle den
 se points. By learning a codebook including common structural features fro
 m simple and sparse shapes, 3QNet can efficiently deal with multiple kinds
  of point clouds. According to experiments on object models, indoor scenes
 , and outdoor scans, 3QNet can achieve better compression performances tha
 n many representative methods.\n\nRegistration Category: FULL ACCESS, ON-D
 EMAND ACCESS\n\nLanguage: ENGLISH\n\nFormat: IN-PERSON, ON-DEMAND
URL:https://sa2022.siggraph.org/en/full-program/?id=papers_327&sess=sess15
 6
END:VEVENT
END:VCALENDAR