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

Guo, Chunle; Li, Chongyi; Guo, Jun; Loy, Chen Change; Hou, Junhui; Kwong, Sam; Cong, Runmin

doi:10.1109/cvpr42600.2020.00185

Cited by 1,307 publications

(1,011 citation statements)

References 30 publications

Supporting

Mentioning

1,003

Contrasting

Unclassified

Order By: Relevance

“…EnlightenGAN [26] proposes a highly effective unsupervised GAN for low-light enhancement. Zero-DCE [19] formulates light enhancement as an imagespecific curve estimation task with a deep network and does not require any paired or unpaired data when training. However, satisfactory results cannot be achieved without paired supervision.…”

Section: Learning-based Methodsmentioning

confidence: 99%

“…Most existing CNNs include the following structures: (1) the resolution of each feature map is the same [19,30,31];…”

Section: Network Structurementioning

confidence: 99%

“…We compared MARN to several state-of-the-art low-light enhancement methods, including SRIE [22], Retinex-Net [16], MBLLEN [25], KinD [17], and Zero-DCE [19]. The results were reproduced using the publicly available source codes provided by the authors.…”

Section: B Evaluation On Benchmark Datasetsmentioning

confidence: 99%

“…In recent years, deep learning methods have been widely used in computer vision tasks and have achieved significant breakthroughs in multiple fields, such as image denoising [7,8,9], image super-resolution [10,11], object detection [12,13], and image enhancement [14,15,16,17,19]. Convolutional neural networks (CNNs) learn strong priors from large-scale datasets.…”

Section: Introductionmentioning

confidence: 99%

“…GANs consist of a generator and a discriminator; however, it is difficult to train two networks simultaneously. Zero-DCE [19] reformulates the task as an image-specific curve estimation problem and only uses several non-reference loss functions. However, the desired enhanced result cannot be easily obtained when only relying on non-reference loss functions.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

MARN: Multi-Scale Attention Retinex Network for Low-Light Image Enhancement

Zhang

Wang

2021

IEEE Access

View full text Add to dashboard Cite

Images captured in low-light conditions often suffer from bad visibility, e.g., low contrast, lost details, and color distortion, and image enhancement methods can be used to improve the image quality. Previous methods have generally obtained a smooth illumination map to enhance the image but have ignored details, leading to inaccurate illumination estimations. To solve this problem, we propose a multiscale attention retinex network (MARN) for low-light image enhancement, which learns an image-toillumination mapping to obtain a detailed inverse illumination map inspired by retinex theory. In order to introduce more image priors, we introduce a novel illuminance-attention map to guide the model to characterize varying-lighting areas, which we combine with the low-light image as the model input. MARN consists of a multi-scale attention module and a feature fusion module; the former extracts multi-resolution features with attention-based feature aggregation, while the latter further merges the output features of the previous module with the input. To achieve better visibility, we formulate a novel loss function to synthetically measure the illumination, detail, and colorfulness of the image. Extensive experiments are performed on several benchmark datasets. The results demonstrate that our method outperforms other state-of-the-art methods according to both objective and subjective metrics.INDEX TERMS Low-light image enhancement, deep learning, multi-scale attention retinex network

show abstract

Section: Learning-based Methodsmentioning

confidence: 99%

“…Most existing CNNs include the following structures: (1) the resolution of each feature map is the same [19,30,31];…”

Section: Network Structurementioning

confidence: 99%

Section: B Evaluation On Benchmark Datasetsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

MARN: Multi-Scale Attention Retinex Network for Low-Light Image Enhancement

Zhang

Wang

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

High Dynamic Range Imaging

Abebe

2023

Fundamentals and Applications of Colour Engineering

View full text Add to dashboard Cite

AFF‐Dehazing: Attention‐based feature fusion network for low‐light image Dehazing

Chen

Sheng

et al. 2021

Computer Animation & Virtual

View full text Add to dashboard Cite

Images captured in haze conditions, especially at nighttime with low light, often suffer from degraded visibility, contrasts, and vividness, which makes it difficult to carry out the following vision tasks. In this article, we propose an attention-based feature fusion network (AFF-Dehazing) for low-light image dehazing. Our method decomposes the low-light image dehazing into two task-independent streams containing four modules: image dehazing module, low-light feature extractor module, feature fusion module, and image restoration module. The basic block of these modules is the proposed attention-based residual dense block. Since the dual-branch are used, AFF-Dehazing can avoid learning the mixed degradation all-in-one and enhance the details of low-light haze images. Extensive experiments show that our method surpasses previous state-of-the-art image dehazing methods and low-light enhancement methods by a very large margin both quantitatively and qualitatively. K E Y W O R D S attention mechanism, image dehazing, low-light enhancement INTRODUCTIONHaze is common atmospheric phenomenon and often plagues the qualities of the images acquired by the cameras. While, at night with low light, the images further face severe degradations, including low visibility and intensity blurs. Recently, many efforts have been made to solve the problems of image dehazing and low-light image enhancement. However, existing methods dealt with the problems of image dehazing and low-light enhancement separately. These methods can achieve relatively good performance on corrupted images when removing a certain target type of distortion (i.e., haze or low-light), but they are not effective in removing both types of distortions.

show abstract

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

Cited by 1,307 publications

References 30 publications

MARN: Multi-Scale Attention Retinex Network for Low-Light Image Enhancement

MARN: Multi-Scale Attention Retinex Network for Low-Light Image Enhancement

High Dynamic Range Imaging

AFF‐Dehazing: Attention‐based feature fusion network for low‐light image Dehazing

Contact Info

Product

Resources

About