BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
X-LIC-LOCATION:Asia/Tokyo
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:JST
DTSTART:18871231T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260817T171531Z
LOCATION:Hall B7 (1)\, B Block\, Level 7
DTSTART;TZID=Asia/Tokyo:20241205T104500
DTEND;TZID=Asia/Tokyo:20241205T105600
UID:siggraphasia_SIGGRAPH Asia 2024_sess129_papers_1108@linklings.com
SUMMARY:Millimetric Human Surface Capture in Minutes
DESCRIPTION:Briac Toussaint and Laurence Boissieux (Centre Inria de l’Univ
 ersité Grenoble Alpes); Diego Thomas (Kyushu University); Edmond Boyer (Me
 ta Reality Labs Research); and Jean-Sébastien Franco (LJK, CNRS, Grenoble 
 INP, Université Grenoble Alpes; Centre Inria de l’Université Grenoble Alpe
 s)\n\nDetailed human surface capture from multiple images is an essential 
 component for many 3D production, analysis and transmission tasks. Yet pro
 ducing millimetric precision 3D models in practical time, and actually ver
 ifying their 3D accuracy in a real-world capture context, remain key chall
 enges due to the lack of specific methods and data for these goals. We pro
 pose two complementary contributions to this end. The first one is a highl
 y scalable neural surface radiance field approach able to achieve millimet
 ric precision by construction, while demonstrating high compute and memory
  efficiency. The second one is a novel dataset, MVMannequin, of clothed ma
 nnequin geometry captured with a high resolution hand-held 3D scanner pair
 ed with calibrated multi-view images, that allows to verify the millimetri
 c accuracy claim. Although our approach can produce such highly dense and 
 precise geometry, we show how aggressive sparsification and optimizations 
 of the neural surface pipeline allow estimations in minutes of computation
  time using only a few GB of GPU memory, while allowing for real-time mill
 isecond neural rendering. On the basis of our framework and dataset, we sh
 ow that our method achieves submillimetric accuracy and completeness for 7
 7% of the points in less than three minutes of training time, with 68 view
 points.\n\nRegistration Category: Full Access, Full Access Supporter\n\nLa
 nguage Format: English Language\n\nSession Chair: Yuting Ye (Reality Labs 
 Research, Meta; Meta)\n\n
URL:https://asia.siggraph.org/2024/program/?id=papers_1108&sess=sess129
END:VEVENT
END:VCALENDAR