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:20251215T131000
DTEND;TZID=Asia/Hong_Kong:20251215T132000
UID:siggraphasia_SIGGRAPH Asia 2025_sess104_papers_1338@linklings.com
SUMMARY:RCTrans: Transparent Object Reconstruction in Natural Scene via Re
 fractive Correspondence Estimation
DESCRIPTION:Fangzhou Gao, Yuzhen Kang, Lianghao Zhang, Li Wang, Qishen Wan
 g, and Jiawan Zhang (Tianjin University)\n\nTransparent object reconstruct
 ion in an uncontrolled natural scene is a challenging task due to its comp
 lex appearance. Existing methods optimize the object shape with RGB color 
 as supervision, which suffer from locality and ambiguity, and fail to reco
 ver fine details. In this paper, we present RC-Trans, which uses ray-backg
 round correspondence as much more efficient constraints to achieve high-qu
 ality reconstruction, while maintaining a convenient setup. The key techno
 logy to achieve this is a novel pre-trained correspondence estimation netw
 ork, which allows us to acquire correspondence under arbitrary scenes and 
 camera views. In addition, a confidence evaluation is introduced to protec
 t the reconstruction from inaccurate estimated correspondence. Extensive e
 xperiments on both synthetic and real data demonstrate that our method can
  produce much more accurate results, without any extra acquisition burden.
  The code and dataset will be publicly available.\n\nRegistration Category
 : Full Access, Full Access Supporter\n\nSession Chair: Xuejin Chen (Univer
 sity of Science and Technology of China)\n\n
END:VEVENT
END:VCALENDAR