Does Haze Removal Help CNN-Based Image Classification?

Pei, Yanting; Huang, Yaping; Zou, Qi; Lu, Yuhang; Wang, Song

doi:10.1007/978-3-030-01249-6_42

Cited by 70 publications

(22 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Perhaps surprisingly at the first glance, we find that almost all existing deraining algorithms will deteriorate the detection performance compared to directly using the rainy images 4 , for YOLO-V3, SSD-512, and Reti-naNet. Our observation concurs the conclusion of another recent study (on dehazing) [51]: since those deraining algorithms were not trained/optimized towards the end goal of object detection, they are unnecessary to help this goal, and the deraining process itself might have lost discriminative, semantically meaningful true information.…”

Section: Task-driven Comparisonsupporting

confidence: 91%

Single Image Deraining: A Comprehensive Benchmark Analysis

Liu

Cao

Araujo

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

268

161

View full text Add to dashboard Cite

Numerous single image deraining algorithms have been recently proposed. However, these algorithms are mainly evaluated using certain type of synthetic images, assuming a specific rain model, plus a few real images. It is thus unclear how these algorithms would perform on rainy images acquired "in the wild" and how we could gauge the progress in the field. This paper aims to bridge this gap. We present a comprehensive study and evaluation of existing single image deraining algorithms, using a new largescale benchmark consisting of both synthetic and realworld rainy images of various rain types. This dataset highlights diverse rain models (rain streak, rain drop, rain and mist), as well as a rich variety of evaluation criteria (fulland no-reference objective, subjective, and task-specific) Our evaluation and analysis indicate the performance gap between synthetic rainy images and real-world images and allow us to better identify the strengths and limitations of each method as well as future research directions.

show abstract

Section: Task-driven Comparisonsupporting

confidence: 91%

Single Image Deraining: A Comprehensive Benchmark Analysis

Liu

Cao

Araujo

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

268

161

View full text Add to dashboard Cite

show abstract

“…Perhaps surprisingly at the first glance, we find that almost all existing deraining algorithms will deteriorate the detection performance compared to directly using the rainy images, for YOLO-V3, SSD-512, and RetinaNet. Our observation concurs the conclusion of another recent study (on dehazing) (Pei et al 2018): since those deraining algo-rithms were not trained/optimized towards the end goal of object detection, they are unnecessary to help this goal, and the deraining process itself might have lost discriminative, semantically meaningful true information.…”

Section: Task-driven Comparisonsupporting

confidence: 91%

A Comprehensive Benchmark Analysis of Single Image Deraining: Current Challenges and Future Perspectives

et al. 2021

View full text Add to dashboard Cite

The capability of image deraining is a highly desirable component of intelligent decision-making in autonomous driving and outdoor surveillance systems. Image deraining aims to restore the clean scene from the degraded image captured in a rainy day. Although numerous single image deraining algorithms have been recently proposed, these algorithms are mainly evaluated using certain type of synthetic images, assuming a specific rain model, plus a few real images. It remains unclear how these algorithms would perform on rainy images acquired "in the wild" and how we could gauge the progress in the field. This paper aims to bridge this gap. We present a comprehensive study and evaluation of existing single image deraining algorithms, using a new large-scale benchmark consisting of both synthetic and real-world rainy images of various rain types. This dataset highlights diverse rain models (rain streak, rain drop, rain and mist), as well as a rich variety of evaluation criteria (full-and no-reference objective, subjective, and task-specific). We further provide a comprehensive suite of criteria for deraining algorithm evaluation, including full-and no-reference metrics, subjective evaluation, and the novel task-driven evaluation. The proposed benchmark is accompanied with extensive experimental results that facilitate the assessment of the state-of-the-arts on a quantitative basis. Our evaluation and analysis indicate the gap between the achievable performance on synthetic rainy images and the practical demand on real-world images. We show that, despite many advances, image deraining is still a largely open problem. The paper is concluded by summarizing our general observations, identifying open research challenges and pointing out future directions. Our code and dataset is publicly available at http://uee.me/ddQsw. Keywords Image deraining • Image quality assessment • Deep convolution network • Benchmark analysis Communicated by Torsten Sattler.

show abstract

“…The object detection results are shown in Table 11, from which we see that only a few detection results are higher than the original detection results. That is because SID methods are not optimized towards the task of object detection, and the deraining process might lose some discriminative and meaningful semantic information [116]. We also calculate the scores of the datasets according to the improvement proportion of the detection results after rain removal, as illustrated in Fig.…”

Section: Test-4: High-level Task Evaluation On Real Datasetmentioning

confidence: 99%

Data-Driven Single Image Deraining: A Comprehensive Review and New Perspectives

Zhang¹,

Yue²,

Zhang³

et al. 2023

Preprint

View full text Add to dashboard Cite

<div>Abstract—Due to the powerful fitting ability of neural networks and massive training data, data-driven Single Image Deraining (SID) methods have obtained significant performance, and most of the existing studies focus on improving the deraining performance by proposing different kinds of deraining networks. However, the generalization ability of current SID methods may still be limited in the real scenario, and the deraining results also cannot effectively improve the performance of subsequent high-level tasks (such as object detection). The main reason may be because the current works mainly focus on designing new deraining neural networks, while neglecting the interpretation of the solving process. To investigate these issues, we in this paper re-examine the three important factors (i.e.,data, rain model and network architecture) for the SID task, and specifically analyze them by proposing new and more reasonable criteria (i.e., general vs. specific, synthetical vs. mathematical, black-box vs. white-box). We will also study the relationship of the three factors from new perspectives of data, and reveal two different solving paradigms (explicit vs. implicit) for the SID task. Besides, we also profoundly studied one of the three factors, i.e., data, and measured the performance of current methods on different datasets via extensive experiments to reveal the effectiveness of SID data. Finally, with the above comprehensive review and in-depth analysis, we also draw some valuable conclusions and suggestions for future research.</div>

show abstract

Does Haze Removal Help CNN-Based Image Classification?

Cited by 70 publications

References 29 publications

Single Image Deraining: A Comprehensive Benchmark Analysis

Single Image Deraining: A Comprehensive Benchmark Analysis

A Comprehensive Benchmark Analysis of Single Image Deraining: Current Challenges and Future Perspectives

Data-Driven Single Image Deraining: A Comprehensive Review and New Perspectives

Contact Info

Product

Resources

About