Retinex-inspired Unrolling with Cooperative Prior Architecture Search for Low-light Image Enhancement

Liu, Risheng; Ma, Long; Zhang, Jiaao; Fan, Xin; Luo, Zhongxuan

doi:10.1109/cvpr46437.2021.01042

Cited by 513 publications

(232 citation statements)

References 32 publications

Supporting

Mentioning

230

Contrasting

Order By: Relevance

“…LIME [10] first estimated the illumination by a prior hypothesis, obtained the estimated illumination through a weighted vibration model, subsequently used BM3D [11] as post-processing. Recently, Liu et al [12] proposed a Retinex-inspired Unrolling with Architecture Search (RUAS) and designed a cooperative reference-free learning strategy to discover low-light prior architectures from a compact search space. Deep learning has been widely used in the field of computer vision and achieves excellent results.…”

Section: A Contrast Enhancement Methodsmentioning

confidence: 99%

R2RNet: Low-light Image Enhancement via Real-low to Real-normal Network

Hai¹,

Zhu²,

Yang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Images captured in weak illumination conditions will seriously degrade the image quality. Solving a series of degradation of low-light images can effectively improve the visual quality of the image and the performance of high-level visual tasks. In this paper, we propose a novel Real-low to Real-normal Network for low-light image enhancement, dubbed R2RNet, based on the Retinex theory, which includes three subnets: a Decom-Net, a Denoise-Net, and a Relight-Net. These three subnets are used for decomposing, denoising, and contrast enhancement, respectively. Unlike most previous methods trained on synthetic images, we collect the first Large-Scale Real-World paired low/normal-light images dataset (LSRW dataset) for training. Our method can properly improve the contrast and suppress noise simultaneously. Extensive experiments on publicly available datasets demonstrate that our method outperforms the existing state-of-the-art methods by a large margin both quantitatively and visually. And we also show that the performance of the highlevel visual task (i.e. face detection) can be effectively improved by using the enhanced results obtained by our method in lowlight conditions. Our codes and the LSRW dataset are available at: https://github.com/abcdef2000/R2RNet.

show abstract

Section: A Contrast Enhancement Methodsmentioning

confidence: 99%

R2RNet: Low-light Image Enhancement via Real-low to Real-normal Network

Hai¹,

Zhu²,

Yang³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The first CNN model LL-Net [21] employs an autoencoder to learn denoising and light enhancement simultaneously. Inspired by the Retinex theory, several LLE networks [20,34,38,40,49] are proposed. They commonly split a low-light input into reflectance and illumination maps, then adjust the illumination map to enhance the intensity.…”

Section: Related Workmentioning

confidence: 99%

“…Enhancement → Deblurring. We choose the recent representative light enhancement networks Zero-DCE [8] and RUAS [20] followed by a state-of-the-art deblurring network MIMO-UNet [6].…”

Section: Evaluation On Lol-blur Datasetmentioning

confidence: 99%

“…As a result, a forceful combination cannot solve the joint degradation caused by low light and motion blur. Specifically, existing low-light enhancement methods [20,38] perform intensity boosting and denoising, ignoring spatial degradation of motion blurs. Instead, motion blur is even enlarged in saturated regions due to over-exposing after performing light enhancement, as shown in Figure 1(b).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

LEDNet: Joint Low-light Enhancement and Deblurring in the Dark

Zhou¹,

Li²,

Loy³

2022

Preprint

View full text Add to dashboard Cite

sg https://shangchenzhou.com/projects/LEDNet (e) LEDNet (Ours) (a) Input (b) RUAS (c) RUAS → MIMO-UNet (d) MIMO-UNet Figure 1. A comparison on the real-world night blurry images shows that existing low-light enhancement and deblurring methods fail in coping with the night blurry images. (a) Input images. (b) Motion blur in the saturated area is enlarged after performing light enhancement using a contemporary method RUAS [20] (indicated by red arrows). (c) Applying a deblurring network MIMO-UNet [6] after light enhancement still fails in blur removal. (d) MIMO-UNet trained on day-time GoPro dataset fails to remove blur in the night-time images. (e) The proposed LEDNet trained with our LOL-Blur dataset yields satisfactory results through joint low-light enhancement and deblurring.

show abstract

“…Fortunately, the learning-based methods based on CNNs recently have been successfully applied to low-light image enhancement task [6]. And they almost surpass the conventional prior-based enhancement methods [1,7,8,9,10] with incredibly good performance.…”

Section: Image Enhancementmentioning

confidence: 99%

Half Wavelet Attention on M-Net+ for Low-Light Image Enhancement

Fan¹,

Liu²,

Liu³

2022

Preprint

View full text Add to dashboard Cite

Low-Light Image Enhancement is a computer vision task which intensifies the dark images to appropriate brightness. It can also be seen as an ill-posed problem in image restoration domain. With the success of deep neural networks, the convolutional neural networks surpass the traditional algorithm-based methods and become the mainstream in the computer vision area. To advance the performance of enhancement algorithms, we propose an image enhancement network (HWMNet) based on an improved hierarchical model: M-Net+. Specifically, we use a half wavelet attention block on M-Net+ to enrich the features from wavelet domain. Furthermore, our HWMNet has competitive performance results on two image enhancement datasets in terms of quantitative metrics and visual quality. The source code and pretrained model are available at https://github.com/FanChiMao/HWMNet.

show abstract

Retinex-inspired Unrolling with Cooperative Prior Architecture Search for Low-light Image Enhancement

Cited by 513 publications

References 32 publications

R2RNet: Low-light Image Enhancement via Real-low to Real-normal Network

R2RNet: Low-light Image Enhancement via Real-low to Real-normal Network

LEDNet: Joint Low-light Enhancement and Deblurring in the Dark

Half Wavelet Attention on M-Net+ for Low-Light Image Enhancement

Contact Info

Product

Resources

About