An efficient deep convolutional laplacian pyramid architecture for CS reconstruction at low sampling ratios

TL;DR

This paper introduces LapCSNet, a deep convolutional Laplacian pyramid architecture for compressed sensing that improves reconstruction quality at low sampling ratios while reducing computational costs.

Contribution

The paper presents a novel deep convolutional Laplacian pyramid network with jointly optimized sampling and reconstruction for enhanced CS image recovery.

Findings

Outperforms state-of-the-art methods in detail preservation.

Reduces computational cost significantly.

Achieves sharper edges and better multi-scale reconstruction.

Abstract

The compressed sensing (CS) has been successfully applied to image compression in the past few years as most image signals are sparse in a certain domain. Several CS reconstruction models have been proposed and obtained superior performance. However, these methods suffer from blocking artifacts or ringing effects at low sampling ratios in most cases. To address this problem, we propose a deep convolutional Laplacian Pyramid Compressed Sensing Network (LapCSNet) for CS, which consists of a sampling sub-network and a reconstruction sub-network. In the sampling sub-network, we utilize a convolutional layer to mimic the sampling operator. In contrast to the fixed sampling matrices used in traditional CS methods, the filters used in our convolutional layer are jointly optimized with the reconstruction sub-network. In the reconstruction sub-network, two branches are designed to reconstruct multi-scale residual images and muti-scale target images progressively using a Laplacian pyramid architecture. The proposed LapCSNet not only integrates multi-scale information to achieve better performance but also reduces computational cost dramatically. Experimental results on benchmark datasets demonstrate that the proposed method is capable of reconstructing more details and sharper edges against the state-of-the-arts methods.