MTS-CSNet: Multiscale Tensor Factorization for Deep Compressive Sensing on RGB Images

TL;DR

MTS-CSNet introduces a multiscale tensor factorization approach for deep compressive sensing of RGB images, enabling large receptive fields and improved reconstruction performance without iterative optimization.

Contribution

The paper proposes MTSCSNet, a novel multiscale tensor summation operator for efficient, non-iterative deep compressive sensing of high-dimensional RGB images.

Findings

Achieves state-of-the-art PSNR on standard benchmarks.

Faster inference compared to diffusion-based methods.

Uses a compact, parameter-efficient architecture.

Abstract

Deep learning based compressive sensing (CS) methods typically learn sampling operators using convolutional or block wise fully connected layers, which limit receptive fields and scale poorly for high dimensional data. We propose MTSCSNet, a CS framework based on Multiscale Tensor Summation (MTS) factorization, a structured operator for efficient multidimensional signal processing. MTS performs mode-wise linear transformations with multiscale summation, enabling large receptive fields and effective modeling of cross-dimensional correlations. In MTSCSNet, MTS is first used as a learnable CS operator that performs linear dimensionality reduction in tensor space, with its adjoint defining the initial back-projection, and is then applied in the reconstruction stage to directly refine this estimate. This results in a simple feed-forward architecture without iterative or proximal optimization, while remaining parameter and computation efficient. Experiments on standard CS benchmarks show that MTSCSNet achieves state-of-the-art reconstruction performance on RGB images, with notable PSNR gains and faster inference, even compared to recent diffusion-based CS methods, while using a significantly more compact feed-forward architecture.