Variational image compression with a scale hyperprior

TL;DR

This paper introduces a variational autoencoder-based image compression model with a hyperprior that captures spatial dependencies, achieving state-of-the-art results in visual quality and rate-distortion performance.

Contribution

It presents a novel end-to-end trainable model incorporating a hyperprior for improved image compression, a concept underexplored in neural network-based codecs.

Findings

Achieves state-of-the-art MS-SSIM quality.

Surpasses existing neural network methods in PSNR-based rate-distortion.

Provides qualitative comparisons across different distortion metrics.

Abstract

We describe an end-to-end trainable model for image compression based on variational autoencoders. The model incorporates a hyperprior to effectively capture spatial dependencies in the latent representation. This hyperprior relates to side information, a concept universal to virtually all modern image codecs, but largely unexplored in image compression using artificial neural networks (ANNs). Unlike existing autoencoder compression methods, our model trains a complex prior jointly with the underlying autoencoder. We demonstrate that this model leads to state-of-the-art image compression when measuring visual quality using the popular MS-SSIM index, and yields rate-distortion performance surpassing published ANN-based methods when evaluated using a more traditional metric based on squared error (PSNR). Furthermore, we provide a qualitative comparison of models trained for different distortion metrics.