DRCNet: Dynamic Image Restoration Contrastive Network

Fei Li, Lingfeng Shen, YANG MI, Zhenbo Li

Abstract:

Image restoration aims to recover images from spatially-varying degradation. Most existing image-restoration models employed
static CNN-based models, where the fixed learned filters cannot fit the diverse degradation well. To this end, we propose a novel Dynamic Image Restoration Contrastive Network (DRCNet) to address this issue. The principal block in DRCNet is the Dynamic Filter Restoration module (DFR), which mainly consists of the spatial filter branch and the energy-based attention branch. Specifically, the spatial filter branch suppresses spatial noise for varying spatial degradation; the energy-based attention branch guides the feature integration for better spatial detail recovery. To make degraded images and clean images more distinctive in the representation space, we develop a novel Intra-class Contrastive Regularization (Intra-CR) to serve as a constraint in the solution space for DRCNet. Meanwhile, our theoretical derivation proved Intra-CR owns less sensitivity towards hyper-parameter selection than previous contrastive regularization. DRCNet outperforms previous methods on the ten widely used benchmarks in image restoration. Besides, the ablation studies investigate the impact of the DFR module and Intra-CR, respectively.