Abstract | ||
---|---|---|
Recovering natural illumination from a single Low-Dynamic Range (LDR) image is a challenging task. To remedy this situation we exploit two properties often found in everyday images. First, images rarely show a single material, but rather multiple ones that all reflect the same illumination. However, the appearance of each material is observed only for some surface orientations, not all. Second, parts of the illumination are often directly observed in the background, without being affected by reflection. Typically, this directly observed part of the illumination is even smaller. We propose a deep Convolutional Neural Network (CNN) that combines prior knowledge about the statistics of illumination and reflectance with an input that makes explicit use of these two observations. Our approach maps multiple partial LDR material observations represented as reflectance maps and a background image to a spherical High-Dynamic Range (HDR) illumination map. For training and testing we propose a new data set comprising of synthetic and real images with multiple materials observed under the same illumination. Qualitative and quantitative evidence shows how both multi-material and using a background are essential to improve illumination estimations. |
Year | Venue | Field |
---|---|---|
2016 | arXiv: Computer Vision and Pattern Recognition | Computer vision,Pattern recognition,Computer science,Convolutional neural network,Exploit,Artificial intelligence,Deep learning,Real image,Reflectivity |
DocType | Volume | Citations |
Journal | abs/1611.09325 | 0 |
PageRank | References | Authors |
0.34 | 0 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Stamatios Georgoulis | 1 | 109 | 10.21 |
Konstantinos Rematas | 2 | 108 | 8.41 |
Tobias Ritschel | 3 | 1052 | 66.60 |
Fritz, Mario | 4 | 2250 | 140.82 |
Tinne Tuytelaars | 5 | 10161 | 609.66 |
Luc Van Gool | 6 | 27566 | 1819.51 |