Curricular Contrastive Regularization for Physics-aware Single Image Dehazing

TL;DR

This paper introduces C2PNet, a physics-aware dehazing network using curricular contrastive regularization and a dual-branch unit, significantly improving single image dehazing performance by better constraining the solution space and enhancing interpretability.

Contribution

The paper proposes a novel curricular contrastive regularization and a physics-aware dual-branch unit, advancing the interpretability and effectiveness of single image dehazing models.

Findings

Achieved PSNR improvements of 3.94dB on SOTS-indoor and 1.50dB on SOTS-outdoor datasets.

Demonstrated superior performance over state-of-the-art methods in dehazing.

Enhanced interpretability through physics-aware feature space modeling.

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 nonconsensual, 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 C2PNet. Extensive experiments demonstrate that our C2PNet 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.