End-to-end Optimized Image Compression

TL;DR

This paper introduces an end-to-end optimized neural image compression method that outperforms traditional codecs in both rate-distortion efficiency and visual quality, inspired by biological neuron models.

Contribution

It presents a novel neural network-based compression framework with joint optimization and local gain control, achieving superior performance over standard methods.

Findings

Better rate-distortion performance than JPEG and JPEG 2000.

Significant visual quality improvements across all bit rates.

Objective quality estimates confirm enhanced image fidelity.

Abstract

We describe an image compression method, consisting of a nonlinear analysis transformation, a uniform quantizer, and a nonlinear synthesis transformation. The transforms are constructed in three successive stages of convolutional linear filters and nonlinear activation functions. Unlike most convolutional neural networks, the joint nonlinearity is chosen to implement a form of local gain control, inspired by those used to model biological neurons. Using a variant of stochastic gradient descent, we jointly optimize the entire model for rate-distortion performance over a database of training images, introducing a continuous proxy for the discontinuous loss function arising from the quantizer. Under certain conditions, the relaxed loss function may be interpreted as the log likelihood of a generative model, as implemented by a variational autoencoder. Unlike these models, however, the compression model must operate at any given point along the rate-distortion curve, as specified by a trade-off parameter. Across an independent set of test images, we find that the optimized method generally exhibits better rate-distortion performance than the standard JPEG and JPEG 2000 compression methods. More importantly, we observe a dramatic improvement in visual quality for all images at all bit rates, which is supported by objective quality estimates using MS-SSIM.