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:20250110T023302Z
LOCATION:G602\, G Block\, Level 6
DTSTART;TZID=Asia/Tokyo:20241205T110000
DTEND;TZID=Asia/Tokyo:20241205T124500
UID:siggraphasia_SIGGRAPH Asia 2024_sess221_crs_116@linklings.com
SUMMARY:Social Presence with Codec Avatars
DESCRIPTION:Courses\n\nJulieta Martinez and Javier Romero (Reality Labs Re
 search) and Shunsuke Saito (Meta)\n\nOur course is devoted to telepresence
 : the tasks of (a) generating realistic human representations with a “deco
 der”, and (b) driving (animating with your body) them with an “encoder”.\n
 \nRegarding generation, we will cover the process of converting a set of h
 igh quality captures into realistic digital human bodies, from data collec
 tion and processing to model learning and rendering.\n\nWe will also cover
  the often overlooked part of _driving_ those models from a VR headset. Th
 is process includes capturing and processing images from headset cameras, 
 generating synthetic data that links those incomplete images to the comple
 te state of the person, and training a model that extracts the driving sig
 nal required to drive the decoder.\n\nA key limiting factor to the develop
 ment of telepresence experiences is a large corpus of high-quality human d
 ata to feed learning algorithms, which is particularly lacking in academia
 . In this regard, we will present two new datasets: Avatar-256, composed o
 f millions of calibrated images of 256 subjects captured in a high resolut
 ion dome for which, besides, every subject has also been captured from a h
 ead-mounted display, allowing researchers in academia to build their own e
 ncoders and decoders for telepresence for the first time. The second datas
 et, Goliath-4, features captures of the same subjects under different cond
 itions (e.g. casual and tight-fitted clothing) captured with different dev
 ices (relightable low volume light stage, large volume photometric scanner
 , smartphone, and head-mounted devices), enabling research into full-body 
 telepresence. We will also present open implementations of baseline, yet s
 tate-of-the-art algorithms that the audience can use to start working on t
 hese challenging problems. We believe that an event in which the science a
 nd data related to realistic telepresence are released to the public will 
 help foster a vibrant research community beyond the few companies that can
  work on this problem at the moment.\n\nRegistration Category: Full Access
 , Full Access Supporter\n\nLanguage Format: English Language
URL:https://asia.siggraph.org/2024/program/?id=crs_116&sess=sess221
END:VEVENT
END:VCALENDAR