Abstract:Block compressed sensing (BCS) is a promising method for resource-constrained image/video coding applications. However, the quantization of BCS measurements has posed a challenge, leading to significant quantization errors and encoding redundancy. In this paper, we propose a quantization method for BCS measurements using convolutional neural networks (CNN). The quantization process maps measurements to quantized data that follow a uniform distribution based on the measurements' distribution, which aims to maximize the amount of information carried by the quantized data. The dequantization process restores the quantized data to data that conform to the measurements' distribution. The restored data are then modified by the correlation information of the measurements drawn from the quantized data, with the goal of minimizing the quantization errors. The proposed method uses CNNs to construct quantization and dequantization processes, and the networks are trained jointly. The distribution parameters of each block are used as side information, which is quantized with 1 bit by the same method. Extensive experiments on four public datasets showed that, compared with uniform quantization and entropy coding, the proposed method can improve the PSNR by an average of 0.48 dB without using entropy coding when the compression bit rate is 0.1 bpp.

Block Compressed Sensing Using Random Permutation and Reweighted Sampling for Image Compression Applications

Optimal Permutation Based Block Compressed Sensing for Image Compression Applications.

Image Reconstruction Via Small Block-Based Compressed Sensing and Discrete Legendre Polynomials

Block Compressed Sensing Based on Image Complexity

Adaptive reweighted compressed sensing for image compression

Random Permutation-Based Block Compressed Sensing For Image Encryptionthen-Compression Applications

Block Compressed Sensing Using Two-dimensional Random Permutation for Image Encryption-then-Compression Applications

Asymmetric Block Based Compressive Sensing for Image Signals

Interweaving Permutation Meets Block Compressed Sensing

Two-Dimensional Compressed Sensing Using Two-Dimensional Random Permutation for Image Encryption-then-Compression Applications.

A Convolutional Neural Network-Based Quantization Method for Block Compressed Sensing of Images

Adaptive sampling for compressed sensing based image compression

Block compressed sampling of image signals by saliency based adaptive partitioning

A new image reconstruction algorithm of block compressed sensing

Blind Compressive Sensing Using Block Sparsity and Nonlocal Low-Rank Priors.

Image compressive sensing using overlapped block projection and reconstruction

Nonuniform Quantization for Block-Based Compressed Sensing of Images in Differential Pulse-Code Modulation Framework

Effective Image Block Compressed Sensing

AMP-BCS: AMP-based image block compressed sensing with permutation of sparsified DCT coefficients

Reconstruction Algorithm for Block-Based Compressed Sensing Based on Mixed Variational Inequality

The Weight-Block Compressed Sensing and its Application to Image Reconstruction