Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement

The paper presents a novel method, Zero-Reference Deep Curve Estimation(Zero-DCE), which formulates light enhancement as a task of image-specificcurve estimation with a deep network. Our method trains a lightweight deepnetwork, DCE-Net, to estimate pixel-wise and high-order curves for dynamicrange adjustment of a given image. The curve estimation is specially designed,considering pixel value range, monotonicity, and differentiability. Zero-DCE isappealing in its relaxed assumption on reference images, i.e., it does notrequire any paired or unpaired data during training. This is achieved through aset of carefully formulated non-reference loss functions, which implicitlymeasure the enhancement quality and drive the learning of the network. Ourmethod is efficient as image enhancement can be achieved by an intuitive andsimple nonlinear curve mapping. Despite its simplicity, we show that itgeneralizes well to diverse lighting conditions. Extensive experiments onvarious benchmarks demonstrate the advantages of our method overstate-of-the-art methods qualitatively and quantitatively. Furthermore, thepotential benefits of our Zero-DCE to face detection in the dark are discussed.Code and model will be available at https://github.com/Li-Chongyi/Zero-DCE.