Single Image Super-Resolution via Wide-Activation Feature Distillation Network

TL;DR

This paper introduces a new network for improving image resolution by combining features from two paths and enhancing detail quality.

Contribution

The novel wide-activation feature distillation network (WFDN) uses dual-path learning and a gated fusion mechanism for better image super-resolution.

Findings

WFDN outperforms state-of-the-art methods in image reconstruction on four benchmark datasets.

The network produces images with richer textures, realistic lines, and clearer structures.

The feature distillation block enables fast training and low parameter usage.

Abstract

Feature extraction plays a pivotal role in the context of single image super-resolution. Nonetheless, relying on a single feature extraction method often undermines the full potential of feature representation, hampering the model’s overall performance. To tackle this issue, this study introduces the wide-activation feature distillation network (WFDN), which realizes single image super-resolution through dual-path learning. Initially, a dual-path parallel network structure is employed, utilizing a residual network as the backbone and incorporating global residual connections to enhance feature exploitation and expedite network convergence. Subsequently, a feature distillation block is adopted, characterized by fast training speed and a low parameter count. Simultaneously, a wide-activation mechanism is integrated to further enhance the representational capacity of high-frequency features. Lastly, a gated fusion mechanism is introduced to weight the fusion of feature information extracted from the dual branches. This mechanism enhances reconstruction performance while mitigating information redundancy. Extensive experiments demonstrate that the proposed algorithm achieves stable and superior results compared to the state-of-the-art methods, as evidenced by quantitative evaluation metrics tests conducted on four benchmark datasets. Furthermore, our WFDN excels in reconstructing images with richer detailed textures, more realistic lines, and clearer structures, affirming its exceptional superiority and robustness.

Figures