Paper Info
Title
Deep reflectance fields: high-quality facial reflectance field inference from color gradient illumination
Abstract
We present a novel technique to relight images of human faces by learning a model of facial reflectance from a database of 4D reflectance field data of several subjects in a variety of expressions and viewpoints. Using our learned model, a face can be relit in arbitrary illumination environments using only two original images recorded under spherical color gradient illumination. The output of our deep network indicates that the color gradient images contain the information needed to estimate the full 4D reflectance field, including specular reflections and high frequency details. While capturing spherical color gradient illumination still requires a special lighting setup, reduction to just two illumination conditions allows the technique to be applied to dynamic facial performance capture. We show side-by-side comparisons which demonstrate that the proposed system outperforms the state-of-the-art techniques in both realism and speed.
Year
DOI
Venue
2019
10.1145/3306346.3323027
ACM Transactions on Graphics (TOG)
Keywords
Field
DocType
machine learning, reflectance estimation
Computer vision,Motion capture,Field data,Expression (mathematics),Inference,Specular reflection,Artificial intelligence,Reflectivity,Mathematics,Color gradient
Journal
Volume
Issue
ISSN
38
4
0730-0301
Citations 
PageRank 
References 
7
0.43
0
Authors
22
Name
Order
Citations
PageRank
Abhimitra Meka1193.02
Christian Hane228117.03
Rohit Pandey3222.64
Michael Zollhöfer485242.25
Sean Ryan Fanello538021.80
Graham Fyffe635622.50
Adarsh Kowdle758424.77
Xueming Yu823918.65
Jay Busch928323.06
Jason Dourgarian10193.64
Peter Denny1170.77
Sofien Bouaziz1293435.79
Peter Lincoln13241.99
Matt Whalen14172.57
Geoff Harvey1591.14
Jonathan Taylor1633414.52
Shahram Izadi175573285.39
Andrea Tagliasacchi1871631.90
Paul Debevec194955449.10
Christian Theobalt203211159.16
Julien Valentin2116311.11
C. Rhemann22124046.56