BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Hong_Kong
X-LIC-LOCATION:Asia/Hong_Kong
BEGIN:STANDARD
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
TZNAME:HKT
DTSTART:19911015T033000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20251218T030656Z
LOCATION:Meeting Room S421\, Level 4
DTSTART;TZID=Asia/Hong_Kong:20251218T110100
DTEND;TZID=Asia/Hong_Kong:20251218T111200
UID:siggraphasia_SIGGRAPH Asia 2025_sess150_papers_1146@linklings.com
SUMMARY:Light-SQ: Structure-aware Shape Abstraction with Superquadrics for
  Generated Meshes
DESCRIPTION:Yuhan Wang (S-Lab for Advanced Intelligence, Nanyang Technolog
 ical University Singapore); Weikai Chen, Zeyu Hu, Runze Zhang, Yingda Yin,
  Ruoyu Wu, Keyang Luo, Shengju Qian, Yiyan Ma, Hongyi Li, Yuan Gao, Yuhuan
  Zhou, Hao Luo, Wan Wang, Xiaobin Shen, Zhaowei Li, Kuixin Zhu, Chuanlang 
 Hong, Yueyue Wang, Lijie Feng, and Xin Wang (LIGHTSPEED); and Chen Change 
 Loy (S-Lab for Advanced Intelligence, Nanyang Technological University Sin
 gapore)\n\nIn user-generated-content (UGC) applications, non-expert users 
 often rely on image-to-3D generative models to create 3D assets. In this c
 ontext, primitive-based shape abstraction offers a promising solution for 
 UGC scenarios by compressing high-resolution meshes into compact, editable
  representations. Towards this end, effective shape abstraction must there
 fore be structure-aware, characterized by low overlap between primitives, 
 part-aware alignment, and primitive compactness. We present Light-SQ, a no
 vel superquadric-based optimization framework that explicitly emphasizes s
 tructure-awareness from three aspects. (a) We introduce SDF carving to ite
 ratively udpate the target signed distance field, discouraging overlap bet
 ween primitives. (b) We propose a block-regrow-fill strategy guided by str
 ucture-aware volumetric decomposition, enabling structural partitioning to
  drive primitive placement. (c) We implement adaptive residual pruning bas
 ed on SDF update history to surpress over-segmentation and ensure compact 
 results. \nIn addition, Light-SQ supports multiscale fitting, enabling loc
 alized refinement to preserve fine geometric details. To evaluate our meth
 od, we introduce 3DGen-Prim, a benchmark extending 3DGen-Bench with new me
 trics for both reconstruction quality and primitive-level editability. Ext
 ensive experiments demonstrate that Light-SQ enables efficient, high-fidel
 ity, and editable shape abstraction with superquadrics for complex generat
 ed geometry, advancing the feasibility of 3D UGC creation.\n\nRegistration
  Category: Full Access, Full Access Supporter\n\nSession Chair: Xuejin Che
 n (University of Science and Technology of China)\n\n
END:VEVENT
END:VCALENDAR