Non-Local Color Compensation Network for Intrinsic Image Decomposition

Single image-based intrinsic image decomposition attempts to separate one input image into several intrinsic components, which is inherently an under-constrained problem. Some recent works have been proposed to estimate the intrinsic components using encoder-decoder structures. However, they generally lack exploration of the different component-oriented feature constraints and feature selection processes. In this paper, a non-local color compensation network (NCCNet) is proposed. Firstly, the hue and value channels of HSV color space are used as the complementary information for RGB images for the estimation of albedo and shading, respectively. The color space representation serves as an external constraint, which does not require expensive sensors or complicated computations. Secondly, an integrated non-local attention scheme is proposed to describe the relations of non-adjacent regions with a lower computational complexity compared to traditional methods. Then the non-local and local attention are combined to describe correlations among features and used as feature selectors between the encoder and decoder. Thirdly, the mutual constraint between albedo and shading is also explored in the network to further optimize the process. In order to train the network, a unified mutual exclusion loss function is proposed. Extensive experiments are conducted on several popular datasets, and the proposed NCCNet achieves improved performance with comparable computational cost compared to competing methods.

Automated summary

This paper proposes a single image-based intrinsic image decomposition method using encoder-decoder structures, which explores different component-oriented feature constraints and feature selection processes. The non-local color compensation network (NCCNet) is introduced to address the computational issue between hue and value channels, and a non-local attention scheme is proposed to describe the relations of non-adjacent regions. The mutual constraint between albedo and shading is also explored, and a unified mutual exclusion loss function is proposed for training.