Estimation of intrinsic images still remains a challenging task due to weaknesses of ground-truth datasets, which either are too small, or present non-realistic issues. On the other hand, end-to-end deep learning architectures start to achieve interesting results that we believe could be improved if important physical hints were not ignored. In this work we present a twofold framework: (a) a flexible generation of images overcoming some classical dataset problems like larger size jointly with coherent lighting appearance; and (b) a flexible architecture tying physical properties through intrinsic losses. Our proposal is versatile, presents low computation time and achieves state-of-art results.

No abstract

No abstract

We present a method to estimate the direction and color of a scene light source from a single image. Our method is based on two main ideas: (a) we use a new synthetic dataset with strong shadow effects with similar constraints to SID dataset; (b) we define a deep architecture trained on the mentioned dataset to estimate direction and color of the scene light source. Apart from showing a good performance on synthetic images, we additionally propose a preliminary procedure to obtain light positions of the Multi-Illumination dataset, and, in this way, we also prove that our trained model achieves a good performance when it is applied to real scenes.

No abstract