Multi-Scale Deep Compressive Sensing Network

TL;DR

This paper introduces MS-DCSNet, a multi-scale deep learning framework for compressive sensing that improves image reconstruction quality by integrating wavelet transforms and multi-scale sampling and reconstruction.

Contribution

It proposes a novel multi-scale DCS network that jointly learns multi-scale sampling and reconstruction using wavelet transforms, enhancing high-frequency detail recovery.

Findings

Improved reconstruction quality over existing methods.

Effective preservation of high-frequency image details.

Enhanced performance at low sampling rates.

Abstract

With joint learning of sampling and recovery, the deep learning-based compressive sensing (DCS) has shown significant improvement in performance and running time reduction. Its reconstructed image, however, losses high-frequency content especially at low subrates. This happens similarly in the multi-scale sampling scheme which also samples more low-frequency components. In this paper, we propose a multi-scale DCS convolutional neural network (MS-DCSNet) in which we convert image signal using multiple scale-based wavelet transform, then capture it through convolution block by block across scales. The initial reconstructed image is directly recovered from multi-scale measurements. Multi-scale wavelet convolution is utilized to enhance the final reconstruction quality. The network is able to learn both multi-scale sampling and multi-scale reconstruction, thus results in better reconstruction quality.