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:20251218T030413Z
LOCATION:Meeting Room S421\, Level 4
DTSTART;TZID=Asia/Hong_Kong:20251215T131000
DTEND;TZID=Asia/Hong_Kong:20251215T141500
UID:siggraphasia_SIGGRAPH Asia 2025_sess104@linklings.com
SUMMARY:3D Reconstruction & Intelligent Geometry
DESCRIPTION:The Technical Papers program is the heartbeat of SIGGRAPH Asia
 , spotlighting world-class scholarly research at the forefront of computer
  graphics and interactive techniques. For decades, it has been the definit
 ive venue where bold ideas take root, foundational concepts are reimagined
 , and the future of visual computing is shaped.\n\nThis year, we explore n
 ew intersections of algorithms and artistry, automation and authorship, to
 ols and imagination – challenging the very way we design, simulate, visual
 ize, and interact with digital worlds.\n\nPhysFiT: Physical-aware 3D Shape
  Understanding for Finishing Incomplete Assembly\n\nUnderstanding the part
  composition and structure of 3D shapes is crucial for a wide range of 3D 
 applications, including 3D part assembly and 3D assembly completion. Compa
 red to 3D part assembly, 3D assembly completion is more complicated which 
 involves repairing broken or incomplete furniture that m...\n\n\nWeihao Wa
 ng (Tongji University College of Elect5ronic and Information Engineering) 
 and Mingyu You, Hongjun Zhou, and Bin He (Tongji University)\n------------
 ---------\nTopology-Aware Optimization of Gaussian Primitives for Human-Ce
 ntric Volumetric Videos\n\nVolumetric video is emerging as a key medium fo
 r digitizing the dynamic physical world, creating the virtual environments
  with six degrees of freedom to deliver immersive user experiences. Howeve
 r, robustly modeling general dynamic scenes, especially those involving to
 pological changes while maintai...\n\n\nYuheng Jiang (Max Planck Institute
  for Informatics, Saarland Informatics Campus; ShanghaiTech University); C
 hengcheng Guo, Yize Wu, Yu Hong, Shengkun Zhu, and Zhehao Shen (ShanghaiTe
 ch University); Yingliang Zhang (DGene Inc.); Shaohui Jiao and Zhuo Su (By
 teDance Inc.); Lan Xu (ShanghaiTech University); and Marc Habermann and Ch
 ristian Theobalt (Max Planck Institute for Informatics, Saarland Informati
 cs Campus)\n---------------------\nLang3D-XL: Language Embedded 3D Gaussia
 ns for Large-scale Scenes\n\nEmbedding a language field in a 3D representa
 tion enables richer semantic understanding of spatial environments by link
 ing geometry with descriptive meaning. This allows for a more intuitive hu
 man-computer interaction, enabling querying or editing scenes using natura
 l language, and could potentially...\n\n\nShai Krakovsky (Tel Aviv Univers
 ity), Gal Fiebelman and Sagie Benaim (Hebrew University of Jerusalem), and
  Hadar Averbuch-Elor (Cornell University)\n---------------------\nSurface-
 Aware Distilled 3D Semantic Features\n\nMany 3D tasks such as pose alignme
 nt, animation, motion transfer, and 3D reconstruction rely on establishing
  correspondences between 3D shapes. This challenge has recently been appro
 ached by pairwise matching of semantic features from pre-trained vision mo
 dels. However, despite their power, these fe...\n\n\nLukas Uzolas, Elmar E
 isemann, and Petr Kellnhofer (Delft University of Technology)\n-----------
 ----------\nRCTrans: Transparent Object Reconstruction in Natural Scene vi
 a Refractive Correspondence Estimation\n\nTransparent object reconstructio
 n in an uncontrolled natural scene is a challenging task due to its comple
 x appearance. Existing methods optimize the object shape with RGB color as
  supervision, which suffer from locality and ambiguity, and fail to recove
 r fine details. In this paper, we present RC-T...\n\n\nFangzhou Gao, Yuzhe
 n Kang, Lianghao Zhang, Li Wang, Qishen Wang, and Jiawan Zhang (Tianjin Un
 iversity)\n---------------------\nPoissonNet: A Local-Global Approach for 
 Learning on Surfaces\n\nMany network architectures exist for learning on m
 eshes, yet their constructions entail delicate trade‑offs between difficul
 ty learning high-frequency features, insufficient receptive field, sensiti
 vity to discretization, and inefficient computational overhead. Drawing fr
 om classic local-globa...\n\n\nArman Maesumi and Tanish Makadia (Brown Uni
 versity), Thibault Groueix and Vladimir Kim (Adobe Research), Daniel Ritch
 ie (Brown University), and Noam Aigerman (University of Montreal)\n\nRegis
 tration Category: Full Access, Full Access Supporter\n\nSession Chair: Xue
 jin Chen (University of Science and Technology of China)
END:VEVENT
END:VCALENDAR