Learning to Restore Low-Light Images via Decomposition-and-Enhancement

By Ke Xu, Xin Yang, Baocai Yin, Rynson W. H. Lau. Computer Vision and Pattern Recognition, 2020

443 citations recorded by the source corpus.

Low-light images typically suffer from two problems. First, they have low visibility (i.e., small pixel values). Second, noise becomes significant and disrupts the image content, due to low signal-to-noise ratio. Most existing lowlight image enhancement methods, however, learn from noise-negligible datasets. They rely on users having good photographic skills in taking images with low noise. Unfortunately, this is not the case for majority of the low-light images. While concurrently enhancing a low-light image and removing its noise is ill-posed, we observe that noise exhibits different levels of contrast in different frequency layers, and it is much easier to detect noise in the lowfrequency layer than in the high one. Inspired by this observation, we propose a frequency-based decompositionand- enhancement model for low-light image enhancement. Based on this model, we present a novel network that first learns to recover image objects in the low-frequency layer and then enhances high-frequency details based on the recovered image objects. In addition, we have prepared a new low-light image dataset with real noise to facilitate learning. Finally, we have conducted extensive experiments to show that the proposed method outperforms state-of-the-art approaches in enhancing practical noisy low-light images.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/205b0077cced9b1680efef87554b4e73f5b04954

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

1. Risheng Liu
2. Xin Fan
3. Zhongxuan Luo
4. Long Ma
5. Ren Yang
6. Ruixing Wang
7. Jiaao Zhang
8. Jiaya Jia

In Media Technology

1. Risheng Liu
2. Xin Fan
3. Zhongxuan Luo
4. Jiaya Jia
5. Long Ma
6. Xiaogang Xu
7. Ruixing Wang
8. Songchen Han

Citations in the corpus

• Retinex-inspired Unrolling with Cooperative Prior Architecture Search for Low-light Image Enhancement (2020, methodology): “The paper in [37] developed a frequency-based decomposition-and-enhancement network based on the attention to context encoding module.”
• R2RNet: Low-light Image Enhancement via Real-low to Real-normal Network (2021, methodology): “Recently, some methods [6] have designed effective models to perform denoising and contrast enhancement simultaneously and obtain satisfactory results.”
• Toward Fast, Flexible, and Robust Low-Light Image Enhancement (2022, unknown): “…four recently-proposed model-based methods (including LECARM [17], SDD [10], STAR [26]), four advanced supervised learning meth-ods (including RetinexNet [3], KinD [34], FIDE [27], DRBN [30]), and four unsupervised learning methods (including EnGAN [11], SSIENet [33], ZeroDCE [7], and RUAS [14]).”
• SNR-Aware Low-light Image Enhancement (2022, methodology): “To this end, we study the relation between signal and noise in image space by exploring Signal-to-Noise-Ratio MIRNet (ECCV2020) IPT (CVPR2021) A3DLUT (ICCV2021) 3DLUT (TPAMI2020) Sprase (TIP2021) Band (TIP2021) Retinex (CVPR2021) FIDE (CVPR2020) EG (TIP2021) Uformer (2021) Ours (cid:1)(cid:5)(cid:8)”