BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
X-LIC-LOCATION:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260805T205430Z
LOCATION:Hall K
DTSTART;TZID=America/Los_Angeles:20260719T180000
DTEND;TZID=America/Los_Angeles:20260719T204500
UID:siggraph_SIGGRAPH 2026_sess237_papers_1217@linklings.com
SUMMARY:Topologically Consistent Multi-view 3D Head Reconstruction via Coa
 rse-Guided Layered Surface Sampling
DESCRIPTION:Timo Bolkart, Daoye Wang, and Prashanth Chandran (Google)\n\nF
 rom calibrated multi-view images, SHELLS reconstructs 18k-vertex 3D heads 
 in 0.08 seconds. It aggregates DinoV2 features via projective surface-awar
 e feature sampling, allowing a transformer to predict dense semantic meshe
 s 3.5x faster with 88% less GPU memory than state-of-the-art methods. The 
 result is an efficient, globally consistent multi-view reconstruction fram
 ework.\n\nInterest Area: Arts & Design, Gaming & Interactive, New Technolo
 gies, Production & Animation, Research & Education\n\nKeyword: Animation, 
 Art, Artificial Intelligence/Machine Learning, Audio, Augmented Reality, C
 apture/Scanning, Computer Vision, Digital Twins, Display, Dynamics, Educat
 ion, Ethics and Society, Fabrication, Games, Generative AI, Geometry, Hapt
 ics, Hardware, Image Processing, Lighting, Math Foundations and Theory, Mo
 deling, Performance, Physical AI, Pipeline Tools and Work, Real-Time, Rend
 ering, Robotics, Scientific Visualization, Simulation, Spatial Computing, 
 Virtual Reality\n\nRegistration Category: Full Conference Supporter, Full 
 Conference, Experience\n\n
END:VEVENT
END:VCALENDAR
