LOW-LIGHT IMAGE ENHANCEMENT VIA FEATURE RESTORATION

Besides poor visibility, under-exposed images often suffer from severe noise and color distortion. Most existingRetinex-based methods deal with the noise and color distortion via some careful designs to denoising and/or colorcorrection. In this paper, we propose a simple yet effectivenetwork from the perspective of feature map restoration tomitigate such issues without constructing any explicit modules. More concretely, we build an encoder-decoder networkto reconstruct images, while a feature restoration subnet isintroduced to transform the features of low-light images tothose of corresponding clear ones. The enhanced images areconsequently acquired through assembling the restored features by the decoder, in which, the noise and possible colordistortion can be greatly remedied. Extensive experiments on widely-used datasets are conducted to validate the superiority of our design over other state-of-the-art alternatives bothquantitatively and qualitatively.