BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Australia/Melbourne
X-LIC-LOCATION:Australia/Melbourne
BEGIN:DAYLIGHT
TZOFFSETFROM:+1000
TZOFFSETTO:+1100
TZNAME:AEDT
DTSTART:19721003T020000
RRULE:FREQ=YEARLY;BYMONTH=4;BYDAY=1SU
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:19721003T020000
TZOFFSETFROM:+1100
TZOFFSETTO:+1000
TZNAME:AEST
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260114T163631Z
LOCATION:Darling Harbour Theatre\, Level 2 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231212T093000
DTEND;TZID=Australia/Melbourne:20231212T124500
UID:siggraphasia_SIGGRAPH Asia 2023_sess209_papers_902@linklings.com
SUMMARY:FLARE: Fast Learning of Animatable and Relightable Mesh Avatars
DESCRIPTION:Shrisha Bharadwaj (Max Planck Institute for Intelligent System
 s); Yufeng Zheng (ETH Zürich, Max Planck Institute for Intelligent Systems
 ); Otmar Hilliges (ETH Zürich); and Michael Black and Victoria Fernandez A
 brevaya (Max Planck Institute for Intelligent Systems)\n\nOur goal is to e
 fficiently learn personalized animatable 3D head avatars from videos that 
 are geometrically accurate, realistic, relightable, and compatible with cu
 rrent rendering systems. While 3D meshes enable efficient processing and a
 re highly portable, they lack realism in terms of shape and appearance. Ne
 ural representations, on the other hand, are realistic but lack compatibil
 ity and are slow to train and render. Our key insight is that it is possib
 le to efficiently learn high-fidelity 3D mesh representations via differen
 tiable rendering, by exploiting highly-optimized methods from traditional 
 computer graphics and approximating some of the components with neural net
 works. Specifically, we introduce FLARE, a technique that enables fast cre
 ation of animatable and relightable mesh avatars from a single monocular v
 ideo. First, we learn a canonical geometry using a mesh representation, en
 abling efficient differentiable rasterization and straightforward animatio
 n via learned blendshapes and linear blend skinning weights. Second, we fo
 llow physically-based rendering and factor observed colors into intrinsic 
 albedo, roughness, and a neural representation of the illumination, allowi
 ng the learned avatars to be relit in novel scenes. Since our input videos
  are captured on a single device with a narrow field of view, modeling the
  surrounding environment light is non-trivial. Based on the split-sum appr
 oximation for modeling specular reflections, we address this by approximat
 ing the pre-filtered environment map with a multi-layer perceptron (MLP) m
 odulated by the surface roughness, eliminating the need to explicitly mode
 l the light. We demonstrate that our mesh-based avatar formulation, combin
 ed with learned deformation, material and lighting MLPs, produces avatars 
 with high-quality geometry and appearance, while also being efficient to t
 rain and render compared to existing approaches.\n\nRegistration Category:
  Full Access, Enhanced Access, Trade Exhibitor, Experience Hall Exhibitor\
 n\n
URL:https://asia.siggraph.org/2023/full-program?id=papers_902&sess=sess209
END:VEVENT
END:VCALENDAR