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

By Chunle Guo, Chongyi Li, Jichang Guo, Chen Change Loy, Junhui Hou, Sam Kwong, Runmin Cong. Computer Vision and Pattern Recognition, 2020

2,237 citations recorded by the source corpus.

The paper presents a novel method, Zero-Reference Deep Curve Estimation (Zero-DCE), which formulates light enhancement as a task of image-specific curve estimation with a deep network. Our method trains a lightweight deep network, DCE-Net, to estimate pixel-wise and high-order curves for dynamic range adjustment of a given image. The curve estimation is specially designed, considering pixel value range, monotonicity, and differentiability. Zero-DCE is appealing in its relaxed assumption on reference images, i.e., it does not require any paired or unpaired data during training. This is achieved through a set of carefully formulated non-reference loss functions, which implicitly measure the enhancement quality and drive the learning of the network. Our method is efficient as image enhancement can be achieved by an intuitive and simple nonlinear curve mapping. Despite its simplicity, we show that it generalizes well to diverse lighting conditions. Extensive experiments on various benchmarks demonstrate the advantages of our method over state-of-the-art methods qualitatively and quantitatively. Furthermore, the potential benefits of our Zero-DCE to face detection in the dark are discussed.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/a2a784f1f24bd400b9ef69f7e0cf7071530d4866

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Learning a Simple Low-Light Image Enhancer from Paired Low-Light Instances (2023, methodology): “PairLIE is compared with 13 state-of-the-art LIE methods, which can be divided into the following three categories: traditional methods (SDD [9], STAR [34]), supervised approaches (MBLLEN [20], RetinexNet [32], GALDNet [30], KinD [38], DRBN [36], URetinexNet [33]), and unsupervised methods (Zero-DCE”
Underwater Image Enhancement via Medium Transmission-Guided Multi-Color Space Embedding (2021, neutral): “Thus, their performance is still far behind when compared with current deep visual models [13]–[16].”
Low-Light Image Enhancement with Wavelet-Based Diffusion Models (2023, methodology): “…CDEF [Lei et al. 2022], 2) learning-based LLIE methods RetinexNet [Wei et al. 2018], DSLR [Lim and Kim 2020], DRBN [Yang et al. 2020a], Zero-DCE [Guo et al. 2020], MIRNet [Zamir et al. 2020], EnlightenGAN [Jiang et al. 2021], ReLLIE [Zhang et al. 2021], RUAS [Liu et al. 2021], SCI [Ma et al.…”
Toward Fast, Flexible, and Robust Low-Light Image Enhancement (2022, neutral): “The smoothness property of the illumination is a broad consensus in this task [7,34].”
Low-Light Image and Video Enhancement Using Deep Learning: A Survey (2021, methodology): “…(n) RRDNet [29] (o) GT (f) KinD [11] (g) KinD++ [61] (h) TBEFN [20] (i) DSLR [21] (j) EnlightenGAN [26] (k) DRBN [33] (l) ExCNet [27] (m) Zero-DCE [28] (n) RRDNet [29] (o) GT requirements are prohibitive to many researchers, especially for beginners who are new to this area and may not even have…”
Learning to Enhance Low-Light Image via Zero-Reference Deep Curve Estimation (2021, result): “This work is an extension of our earlier conference version that has appeared in CVRP2020[11].”
UIEC^2-Net: CNN-based Underwater Image Enhancement Using Two Color Space (2021, background): “In the past few decades, deep learning technology has developed rapidly, which has brought a series of breakthroughs to various of computer vision and image processing tasks [9]–[16].”
Retinex-inspired Unrolling with Cooperative Prior Architecture Search for Low-light Image Enhancement (2020, methodology): “In addition, dozens of methods also train their networks without paired supervision [40, 12, 8].”
Low-Light Image Enhancement with Normalizing Flow (2021, methodology): “(Ying, Li, and Gao 2017) 15.95 0.6386 0.4573 DeepUPE (Wang et al. 2019a) 13.19 0.4902 0.4634 JED (Ren et al. 2018) 16.73 0.6817 0.3899 LIME (Guo, Li, and Ling 2016) 14.07 0.5274 0.4021 SICE (Cai, Gu, and Zhang 2018) 18.06 0.7094 0.5078 LLNet (Lore et al. 2017) 17.57 0.7388 0.4021 SRIE (Fu et al. 201”
URetinex-Net: Retinex-based Deep Unfolding Network for Low-light Image Enhancement (2022, methodology): “Furthermore, to verify the efficiency of our model in learning implicit prior from data, we compare with state-ofthe-art (SOTA) learning-based methods including RetinexNet [34], KinD [41], Zero-DCE [11], RUAS [24], AGLLNet [26] and KinD++ [40].”
Image-Adaptive YOLO for Object Detection in Adverse Weather Conditions (2021, positive): “Another approach is to dilute the inﬂuence of weather-speciﬁc information by preprocessing the image with existing methods such as image defogging (Hang et al. 2020; Liu et al. 2019) and image enhancement (Guo et al. 2020).”
SNR-Aware Low-light Image Enhancement (2022, background): “Also, there are unsupervised methods [7, 14, 19].”
Ultra-High-Definition Low-Light Image Enhancement: A Benchmark and Transformer-Based Method (2022, methodology): “2021b)), two zero-shot learning-based methods (Z_DCE (Guo et al. 2020), Z_DCE++ (Li, Guo, and Loy 2021) ) and two supervised transformer-based methods (Uformer (Wang et al.”
PUGAN: Physical Model-Guided Underwater Image Enhancement Using GAN With Dual-Discriminators (2023, background): “light scene [5], [6], [7] or foggy scene [8], [9], [10], [11]) do not generalize well to the UIE task.”