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:20251218T030657Z
LOCATION:Meeting Room S423+S424\, Level 4
DTSTART;TZID=Asia/Hong_Kong:20251217T172400
DTEND;TZID=Asia/Hong_Kong:20251217T173500
UID:siggraphasia_SIGGRAPH Asia 2025_sess144_papers_2209@linklings.com
SUMMARY:LookUp3D: Data-Driven 3D Scanning
DESCRIPTION:Giancarlo Pereira, Yidan Gao, Yurii Piadyk, David Fouhey, Clau
 dio Silva, and Daniele Panozzo (New York University)\n\nHigh speed, high-r
 esolution, and accurate 3D scanning would open doors to many new applicati
 ons in graphics, robotics, science, and medicine by enabling the accurate 
 scanning of deformable objects during interactions. Past attempts to use s
 tructured light, time-of-flight, and stereo in high-speed settings have us
 ually required tradeoffs in resolution or inaccuracy. In this paper, we in
 troduce a method that enables, for the first time, 3D scanning at 450 fram
 es per second at 1 Megapixel, or 1,450 frames per second at 0.4 Megapixel 
 in an environment with controlled lighting. The key idea is to use a per-p
 ixel lookup table that maps colors to depths, which is built using a linea
 r stage. Imperfections, such as lens-distortion and sensor defects are bak
 ed into the calibration. We describe our method and test it on a novel har
 dware prototype. We compare the system with both ground-truth geometry as 
 well as commercially available dynamic sensors like the Microsoft Kinect a
 nd Intel Realsense. Our results show the system acquiring geometry of obje
 cts undergoing high-speed deformations and oscillations and demonstrate th
 e ability to recover physical properties from the reconstructions.\n\nRegi
 stration Category: Full Access, Full Access Supporter\n\nSession Chair: Se
 ung-Hwan Baek (POSTECH)\n\n
END:VEVENT
END:VCALENDAR