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

2026-02-04

AI Summary

This work proposes MTS-CSNet, a novel deep compressive sensing framework that overcomes the limitations of conventional methods—namely, restricted receptive fields and poor scalability to high-dimensional data. By introducing a learnable structured compression operator based on Multi-scale Tensor Sum (MTS) decomposition, the model performs linear dimensionality reduction and feedforward reconstruction directly in tensor space, eliminating the need for iterative optimization. This approach effectively captures long-range spatial dependencies and cross-dimensional correlations while maintaining parameter efficiency. Evaluated on standard RGB image compressive sensing tasks, MTS-CSNet achieves state-of-the-art performance in terms of PSNR, offers faster inference than recent diffusion-based methods, and features a more compact architecture.

Technology Category

• Machine Learning: Matrix & Tensor Methods
• Computer Vision: Diffusion Models for Vision
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

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.

Problem

• compressive sensing
• RGB images
• deep learning
• high-dimensional data
• receptive field

Innovation

• Multiscale Tensor Factorization
• Compressive Sensing
• Tensor-based Sampling
• Feed-forward Reconstruction
• Cross-dimensional Correlation