Publication Type

Conference Proceeding Article

Version

publishedVersion

Publication Date

6-2023

Abstract

Considering the ill-posed nature, contrastive regularization has been developed for single image dehazing, introducing the information from negative images as a lower bound. However, the contrastive samples are non-consensual, as the negatives are usually represented distantly from the clear (i.e., positive) image, leaving the solution space still under-constricted. Moreover, the interpretability of deep dehazing models is underexplored towards the physics of the hazing process. In this paper, we propose a novel curricular contrastive regularization targeted at a consensual contrastive space as opposed to a non-consensual one. Our negatives, which provide better lower-bound constraints, can be assembled from 1) the hazy image, and 2) corresponding restorations by other existing methods. Further, due to the different similarities between the embeddings of the clear image and negatives, the learning difficulty of the multiple components is intrinsically imbalanced. To tackle this issue, we customize a curriculum learning strategy to reweight the importance of different negatives. In addition, to improve the interpretability in the feature space, we build a physics-aware dual-branch unit according to the atmospheric scattering model. With the unit, as well as curricular contrastive regularization, we establish our dehazing network, named C 2 PNet. Extensive experiments demonstrate that our C 2 PNet significantly outperforms state-of-the-art methods, with extreme PSNR boosts of 3.94dB and 1.50dB, respectively, on SOTS-indoor and SOTS-outdoor datasets. Code is available at https://github.com/YuZheng9/C2PNet.

Keywords

Low-level vision, Computer vision, Codes, Atmospheric modeling, Computational modeling, Scattering, Pattern recognition, Image restoration

Discipline

Artificial Intelligence and Robotics | Graphics and Human Computer Interfaces

Publication

Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, BC, Canada, June 17-24, 2023

First Page

5785

Last Page

5794

ISBN

9798350301304

Identifier

10.1109/CVPR52729.2023.00560

Publisher

IEEE Computer Society

City or Country

New York, NY, USA

Copyright Owner and License

Authors

Additional URL

https://doi.org/10.1109/CVPR52729.2023.00560

Download

Share

COinS