Paper Info
Title
DeLight-Net: Decomposing Reflectance Maps into Specular Materials and Natural Illumination.
Abstract
In this paper we are extracting surface reflectance and natural environmental illumination from a reflectance map, i.e. from a single 2D image of a sphere of one material under one illumination. This is a notoriously difficult problem, yet key to various re-rendering applications. With the recent advances in estimating reflectance maps from 2D images their further decomposition has become increasingly relevant. To this end, we propose a Convolutional Neural Network (CNN) architecture to reconstruct both material parameters (i.e. Phong) as well as illumination (i.e. high-resolution spherical illumination maps), that is solely trained on synthetic data. We demonstrate that decomposition of synthetic as well as real photographs of reflectance maps, both in High Dynamic Range (HDR), and, for the first time, on Low Dynamic Range (LDR) as well. Results are compared to previous approaches quantitatively as well as qualitatively in terms of re-renderings where illumination, material, view or shape are changed.
Year
Venue
Field
2016
arXiv: Computer Vision and Pattern Recognition
Computer vision,Pattern recognition,Computer science,Convolutional neural network,Specular reflection,Low dynamic range,Synthetic data,Artificial intelligence,Reflectivity,High dynamic range
DocType
Volume
Citations 
Journal
abs/1603.08240
3
PageRank 
References 
Authors
0.38
10
6
Name
Order
Citations
PageRank
Stamatios Georgoulis110910.21
Konstantinos Rematas21088.41
Tobias Ritschel3105266.60
Mario Fritz4211.75
Luc Van Gool5275661819.51
Tinne Tuytelaars610161609.66