Improved Techniques for Learning to Dehaze and Beyond: A Collective Study

Liu, Yu; Zhao, Guanlong; Gong, Boyuan; Li, Yang; Raj, Ritu; Goel, Niraj; Kesav, Satya; Gottimukkala, Sandeep; Wang, Zhangyang; Ren, Wenqi; Tao, Dacheng

doi:10.48550/arxiv.1807.00202

Cited by 14 publications

(22 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…And DMask-RCNN adds a domain-adaptive branch after the base feature extraction layers in Mask R-CNN architecture, aiming to mask the generated features to be domain-invariant between the source domain and target domain. The experimental results in [39] [41] demonstrate that the domain adaptation method can enhance the performance of both Faster R-CNN and Mask R-CNN models when tackling the object detection task in the hazy environment. Moreover, this enhancement can be more effective when feeding the target domain with images restored by a robust dehazing algorithm.…”

Section: Domain Adaption Methodsmentioning

confidence: 96%

“…4.2 Object Detection with Pre-dehazed Test Dataset 4.2.1 Experiment Setup Some object detection tasks on RESIDE RTTS dataset have been tested in [7] [11] [41]. The Domain-Adaptive Mask R-CNN model proposed in [41] achieves the highest detection accuracy on RTTS test dataset among several well-known object detection models including Faster R-CNN [35], Mask R-CNN [13], SSD [45] and [46]. [7] proposed a concatenation of dehazing algorithm and object detection modules to detect objects in the hazy environment.…”

Section: Real-world Hazy Images Dehazingmentioning

confidence: 99%

“…[7] proposed a concatenation of dehazing algorithm and object detection modules to detect objects in the hazy environment. [41] further tried more combinations of effective dehazing algorithms and object detectors in the cascade and evaluated their performances on the mean average precision (mAP) values. MLDCP model is utilized to dehaze the RTTS test dataset, and Mask R-CNN and DMask-RCNN modules are fed with the recovered images.…”

Section: Real-world Hazy Images Dehazingmentioning

confidence: 99%

“…MLDCP model is utilized to dehaze the RTTS test dataset, and Mask R-CNN and DMask-RCNN modules are fed with the recovered images. Then the authors compare the detection accuracy with the results of dehazing-detection cascades in [41].…”

Section: Real-world Hazy Images Dehazingmentioning

confidence: 99%

See 3 more Smart Citations

Advanced Multiple Linear Regression Based Dark Channel Prior Applied on Dehazing Image and Generating Synthetic Haze

Li¹,

Hua²,

Lu³

2021

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

Haze removal is an extremely challenging task, and object detection in the hazy environment has recently gained much attention due to the popularity of autonomous driving and traffic surveillance. In this work, the authors propose a multiple linear regression haze removal model based on a widely adopted dehazing algorithm named Dark Channel Prior. Training this model with a synthetic hazy dataset, the proposed model can reduce the unanticipated deviations generated from the rough estimations of transmission map and atmospheric light in Dark Channel Prior. To increase object detection accuracy in the hazy environment, the authors further present an algorithm to build a synthetic hazy COCO training dataset by generating the artificial haze to the MS COCO training dataset. The experimental results demonstrate that the proposed model obtains higher image quality and shares more similarity with ground truth images than most conventional pixel-based dehazing algorithms and neural network based haze-removal models. The authors also evaluate the mean average precision of Mask R-CNN when training the network with synthetic hazy COCO training dataset and preprocessing test hazy dataset by removing the haze with the proposed dehazing model. It turns out that both approaches can increase the object detection accuracy significantly and outperform most existing object detection models over hazy images.

show abstract

Section: Domain Adaption Methodsmentioning

confidence: 96%

Section: Real-world Hazy Images Dehazingmentioning

confidence: 99%

Section: Real-world Hazy Images Dehazingmentioning

confidence: 99%

Section: Real-world Hazy Images Dehazingmentioning

confidence: 99%

See 2 more Smart Citations

Advanced Multiple Linear Regression Based Dark Channel Prior Applied on Dehazing Image and Generating Synthetic Haze

Li¹,

Hua²,

Lu³

2021

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

show abstract

“…Image enhancement as pre-processing for improving subsequent high-level vision tasks has recently received in- creasing attention [47], [48]. We investigate the impact of light enhancement on the DARK FACE dataset 2 , which was specifically built for the task of face detection in low-light conditions.…”

Section: E Pre-processing For Improving Face Detectionmentioning

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

U-Finger: Multi-Scale Dilated Convolutional Network for Fingerprint Image Denoising and Inpainting

Wang

et al. 2019

Inpainting and Denoising Challenges

Self Cite

View full text Add to dashboard Cite

This paper studies the challenging problem of fingerprint image denoising and inpainting. To tackle the challenge of suppressing complicated artifacts (blur, brightness, contrast, elastic transformation, occlusion, scratch, resolution, rotation, and so on) while preserving fine textures, we develop a multi-scale convolutional network termed U-Finger. Based on the domain expertise, we show that the usage of dilated convolutions as well as the removal of padding have important positive impacts on the final restoration performance, in addition to multi-scale cascaded feature modules. Our model achieves the overall ranking of No.2 in the ECCV 2018 Chalearn LAP Inpainting Competition Track 3 (Fingerprint Denoising and Inpainting). Among all participating teams, we obtain the MSE of 0.0231 (rank 2), PSNR 16.9688 dB (rank 2), and SSIM 0.8093 (rank 3) on the hold-out testing set.

show abstract

Improved Techniques for Learning to Dehaze and Beyond: A Collective Study

Cited by 14 publications

References 23 publications

Advanced Multiple Linear Regression Based Dark Channel Prior Applied on Dehazing Image and Generating Synthetic Haze

Advanced Multiple Linear Regression Based Dark Channel Prior Applied on Dehazing Image and Generating Synthetic Haze

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

U-Finger: Multi-Scale Dilated Convolutional Network for Fingerprint Image Denoising and Inpainting

Contact Info

Product

Resources

About