Localized Compression: Applying Convolutional Neural Networks to Compressed Images

TL;DR

This paper introduces Localized Compression, a method that divides images into blocks and compresses each to improve CNN classification accuracy on compressed images, outperforming traditional downgrading.

Contribution

The paper proposes a novel localized compression technique that aligns block compression with CNN convolutional regions, enhancing classification performance on compressed images.

Findings

Localized Compression achieves 1-2% higher accuracy than downgrading.

The method is compatible with any CNN architecture.

Experimental results validate the effectiveness of localized compression.

Abstract

We address the challenge of applying existing convolutional neural network (CNN) architectures to compressed images. Existing CNN architectures represent images as a matrix of pixel intensities with a specified dimension; this desired dimension is achieved by downgrading or cropping. Downgrading and cropping are attractive in that the result is also an image; however, an algorithm producing an alternative "compressed" representation could yield better classification performance. This compression algorithm need not be reversible, but must be compatible with the CNN's operations. This problem is thus the counterpart of the well-studied problem of applying compressed CNNs to uncompressed images, which has attracted great interest as CNNs are deployed to size-, weight-, and power- (SWaP)-limited devices. We introduce Localized Compression, a generalization of downgrading in which the original image is divided into blocks and each block is compressed to a smaller size using either sampling- or random-matrix-based techniques. By aligning the size of the compressed blocks with the size of the CNN's convolutional region, localized compression can be made compatible with any CNN architecture. Our experimental results show that Localized Compression results in classification accuracy approximately 1-2% higher than is achieved by downgrading to the equivalent resolution.