RetinexDIP: A Unified Deep Framework for Low-Light Image Enhancement

Zhao, Zunjin; Xiong, Bangshu; Wang, Lei; Ou, Qiaofeng; Lei, Yu; Kuang, Fa

doi:10.1109/tcsvt.2021.3073371

Cited by 218 publications

(68 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For example, Above methods were all built in a fully supervised fashion, and there are also some recent studies on unsupervised learning. For example, Zhao et al (2021) built Retinex-DIP model by combining the Retinex model and deep image prior (Ulyanov, Vedaldi, & Lempitsky, 2018); Zhu et al (2020) proposed RRDNet by training the network in a zero-shot way with specifically designed loss functions.…”

Section: Retinex Inspired Deep Learning Methodsmentioning

confidence: 99%

Low-light Image Enhancement by Retinex Based Algorithm Unrolling and Adjustment

Liu¹,

Xie²,

Zhao³

et al. 2022

Preprint

View full text Add to dashboard Cite

Motivated by their recent advances, deep learning techniques have been widely applied to low-light image enhancement (LIE) problem. Among which, Retinex theory based ones, mostly following a decomposition-adjustment pipeline, have taken an important place due to its physical interpretation and promising performance. However, current investigations on Retinex based deep learning are still not sufficient, ignoring many useful experiences from traditional methods. Besides, the adjustment step is either performed with simple image processing techniques, or by complicated networks, both of which are unsatisfactory in practice. To address these issues, we propose a new deep learning framework for the LIE problem. The proposed framework contains a decomposition network inspired by algorithm unrolling, and adjustment networks considering both global brightness and local brightness sensitivity. By virtue of algorithm unrolling, both implicit priors learned from data and explicit priors borrowed from traditional methods can be embedded in the network, facilitate to better decomposition. Meanwhile, the consideration of global and local brightness can guide designing simple yet effective network modules for adjustment. Besides, to avoid manually parameter tuning, we also propose a self-supervised fine-tuning strategy, which can always guarantee a promising performance. Experiments on a series of typical LIE datasets demonstrated the effectiveness of the proposed method, both quantitatively and visually, as compared with existing methods.

show abstract

Section: Retinex Inspired Deep Learning Methodsmentioning

confidence: 99%

Low-light Image Enhancement by Retinex Based Algorithm Unrolling and Adjustment

Liu¹,

Xie²,

Zhao³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Illumination-consistency Loss. As in [ 47 ], we also consider the illumination-consistency loss, which is defined as

where

is the initial illumination obtained by

for every pixel p .…”

Section: Methodsmentioning

confidence: 99%

“…In this section, we show the effectiveness of the proposed method. We compare it with six other methods, i.e., LIME [ 21 ], NPE [ 19 ], SRIE [ 63 ], KinD [ 36 ], Zero-DCE [ 55 ], and RetinexDIP [ 47 ].…”

Section: Methodsmentioning

confidence: 99%

“…Fu et al [ 46 ] suggested a novel unsupervised low-light image enhancement network (LE-GAN) based on generative adversarial networks using unpaired low-light/normal-light images for training. Zhao et al [ 47 ] proposed a unified deep zero-reference framework termed RetinexDIP for enhancing low-light images; however, noise was not considered in the decomposition process. Liu et al [ 48 ] proposed the Retinex-based fast algorithm (RBFA) to achieve low-light image enhancement.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Low-Light Image Enhancement via Retinex-Style Decomposition of Denoised Deep Image Prior

Gao

Zhang

Luo

2022

Sensors

View full text Add to dashboard Cite

Low-light images are a common phenomenon when taking photos in low-light environments with inappropriate camera equipment, leading to shortcomings such as low contrast, color distortion, uneven brightness, and high loss of detail. These shortcomings are not only subjectively annoying but also affect the performance of many computer vision systems. Enhanced low-light images can be better applied to image recognition, object detection and image segmentation. This paper proposes a novel RetinexDIP method to enhance images. Noise is considered as a factor in image decomposition using deep learning generative strategies. The involvement of noise makes the image more real, weakens the coupling relationship between the three components, avoids overfitting, and improves generalization. Extensive experiments demonstrate that our method outperforms existing methods qualitatively and quantitatively.

show abstract

“…SID [46] uses a U-Net to enhance the extremely dark RAW image. RetinexDIP [47] provides a unified deep framework using a novel "generative" strategy for Retinex decomposition. Zhang et al [48] presented a self-supervised low-light image enhancement network, which is only trained with low-light images.…”

Section: Related Workmentioning

confidence: 99%