Towards Image Understanding from Deep Compression without Decoding

TL;DR

This paper demonstrates that performing image classification and segmentation directly on deep compressed representations can achieve comparable accuracy to RGB-based methods while reducing computational costs, especially at high compression rates.

Contribution

It introduces a method to perform image understanding directly on compressed representations, bypassing decoding, and shows joint training improves performance and efficiency.

Findings

Achieves comparable accuracy to RGB-based methods on compressed representations.

Reduces computational complexity by up to 2x.

Joint training enhances image quality and understanding tasks.

Abstract

Motivated by recent work on deep neural network (DNN)-based image compression methods showing potential improvements in image quality, savings in storage, and bandwidth reduction, we propose to perform image understanding tasks such as classification and segmentation directly on the compressed representations produced by these compression methods. Since the encoders and decoders in DNN-based compression methods are neural networks with feature-maps as internal representations of the images, we directly integrate these with architectures for image understanding. This bypasses decoding of the compressed representation into RGB space and reduces computational cost. Our study shows that accuracies comparable to networks that operate on compressed RGB images can be achieved while reducing the computational complexity up to $2\times$. Furthermore, we show that synergies are obtained by jointly training compression networks with classification networks on the compressed representations, improving image quality, classification accuracy, and segmentation performance. We find that inference from compressed representations is particularly advantageous compared to inference from compressed RGB images for aggressive compression rates.