FSOINet: Feature-Space Optimization-Inspired Network for Image Compressive Sensing

TL;DR

FSOINet introduces a novel feature-space optimization approach for image compressive sensing, leveraging learned sampling and phase-wise information flow in feature space, resulting in superior reconstruction performance.

Contribution

The paper proposes FSOINet, a network that maps optimization steps into feature space and learns the sampling matrix end-to-end, enhancing image reconstruction quality.

Findings

Outperforms state-of-the-art methods significantly

Achieves better quantitative and qualitative results

Effectively utilizes feature space for information flow

Abstract

In recent years, deep learning-based image compressive sensing (ICS) methods have achieved brilliant success. Many optimization-inspired networks have been proposed to bring the insights of optimization algorithms into the network structure design and have achieved excellent reconstruction quality with low computational complexity. But they keep the information flow in pixel space as traditional algorithms by updating and transferring the image in pixel space, which does not fully use the information in the image features. In this paper, we propose the idea of achieving information flow phase by phase in feature space and design a Feature-Space Optimization-Inspired Network (dubbed FSOINet) to implement it by mapping both steps of proximal gradient descent algorithm from pixel space to feature space. Moreover, the sampling matrix is learned end-to-end with other network parameters. Experiments show that the proposed FSOINet outperforms the existing state-of-the-art methods by a large margin both quantitatively and qualitatively. The source code is available on https://github.com/cwjjun/FSOINet.