Towards Context-aware Convolutional Network for Image Restoration

TL;DR

This paper introduces CCNet, a novel convolutional network for image restoration that effectively captures long-range context and high-dimensional features, outperforming existing methods across multiple IR tasks.

Contribution

The paper proposes ERSM and LDIM modules integrated into a U-shaped network to enhance context-aware feature mapping and contextual information extraction in image restoration.

Findings

CCNet achieves superior performance on dehazing, deblurring, and desnowing tasks.

The proposed modules significantly improve feature representation and contextual information utilization.

CCNet maintains low model complexity while outperforming state-of-the-art methods.

Abstract

Image restoration (IR) is a long-standing task to recover a high-quality image from its corrupted observation. Recently, transformer-based algorithms and some attention-based convolutional neural networks (CNNs) have presented promising results on several IR tasks. However, existing convolutional residual building modules for IR encounter limited ability to map inputs into high-dimensional and non-linear feature spaces, and their local receptive fields have difficulty in capturing long-range context information like Transformer. Besides, CNN-based attention modules for IR either face static abundant parameters or have limited receptive fields. To address the first issue, we propose an efficient residual star module (ERSM) that includes context-aware "star operation" (element-wise multiplication) to contextually map features into exceedingly high-dimensional and non-linear feature spaces, which greatly enhances representation learning. To further boost the extraction of contextual information, as for the second issue, we propose a large dynamic integration module (LDIM) which possesses an extremely large receptive field. Thus, LDIM can dynamically and efficiently integrate more contextual information that helps to further significantly improve the reconstruction performance. Integrating ERSM and LDIM into an U-shaped backbone, we propose a context-aware convolutional network (CCNet) with powerful learning ability for contextual high-dimensional mapping and abundant contextual information. Extensive experiments show that our CCNet with low model complexity achieves superior performance compared to other state-of-the-art IR methods on several IR tasks, including image dehazing, image motion deblurring, and image desnowing.