Compressed Domain Image Classification Using a Dynamic-Rate Neural Network

TL;DR

This paper introduces a neural network trained with a novel dynamic-rate scheme that can classify images directly from compressive measurements across a range of measurement rates, eliminating the need for multiple models.

Contribution

The authors develop a training method enabling a single neural network to perform classification over various measurement rates in compressed domain imaging.

Findings

Achieves comparable accuracy to single-rate models at multiple measurement rates.

Demonstrates robustness to noise in compressed measurements.

Applicable across different datasets and sensing matrices.

Abstract

Compressed domain image classification performs classification directly on compressive measurements acquired from the single-pixel camera, bypassing the image reconstruction step. It is of great importance for extending high-speed object detection and classification beyond the visible spectrum in a cost-effective manner especially for resource-limited platforms. Previous neural network methods require training a dedicated neural network for each different measurement rate (MR), which is costly in computation and storage. In this work, we develop an efficient training scheme that provides a neural network with dynamic-rate property, where a single neural network is capable of classifying over any MR within the range of interest with a given sensing matrix. This training scheme uses only a few selected MRs for training and the trained neural network is valid over the full range of MRs of interest. We demonstrate the performance of the dynamic-rate neural network on datasets of MNIST, CIFAR-10, Fashion-MNIST, COIL-100, and show that it generates approximately equal performance at each MR as that of a single-rate neural network valid only for one MR. Robustness to noise of the dynamic-rate model is also demonstrated. The dynamic-rate training scheme can be regarded as a general approach compatible with different types of sensing matrices, various neural network architectures, and is a valuable step towards wider adoption of compressive inference techniques and other compressive sensing related tasks via neural networks.