s-LWSR: Super Lightweight Super-Resolution Network

TL;DR

This paper introduces s-LWSR, a super lightweight super-resolution network designed for mobile devices, utilizing an information pool, a compression module, and activation layer removal to reduce parameters while maintaining high performance.

Contribution

The work proposes a novel lightweight SR network with an information pool, a compression module, and activation layer removal, enabling efficient super-resolution on mobile devices.

Findings

Achieves comparable performance to larger models with fewer parameters.

Uses an information pool to effectively combine multi-level features.

Demonstrates the effectiveness of activation removal for model efficiency.

Abstract

Deep learning (DL) architectures for superresolution (SR) normally contain tremendous parameters, which has been regarded as the crucial advantage for obtaining satisfying performance. However, with the widespread use of mobile phones for taking and retouching photos, this character greatly hampers the deployment of DL-SR models on the mobile devices. To address this problem, in this paper, we propose a super lightweight SR network: s-LWSR. There are mainly three contributions in our work. Firstly, in order to efficiently abstract features from the low resolution image, we build an information pool to mix multi-level information from the first half part of the pipeline. Accordingly, the information pool feeds the second half part with the combination of hierarchical features from the previous layers. Secondly, we employ a compression module to further decrease the size of parameters. Intensive analysis confirms its capacity of trade-off between model complexity and accuracy. Thirdly, by revealing the specific role of activation in deep models, we remove several activation layers in our SR model to retain more information for performance improvement. Extensive experiments show that our s-LWSR, with limited parameters and operations, can achieve similar performance to other cumbersome DL-SR methods.