Quantum mechanics

Computer hardware

Algorithm

Residual

Image compression

Context model

Entropy encoding

Entropy (arrow of time)

Time series

Image processing

Telecommunications

Computer science

STAR model

Codec

Physics

Pattern recognition (psychology)

Mathematics

Image (mathematics)

Autoregressive model

Object (grammar)

Autoencoder

Econometrics

Autoregressive integrated moving average

Artificial intelligence

Data compression

Channel (broadcasting)

Artificial neural network

Machine learning

Speech recognition

In learning-based approaches to image compression, codecs are developed by optimizing a computational model to minimize a rate-distortion objective. Currently, the most effective learned image codecs take the form of an entropy-constrained autoencoder with an entropy model that uses both forward and backward adaptation. Forward adaptation makes use of side information and can be efficiently integrated into a deep neural network. In contrast, backward adaptation typically makes predictions based on the causal context of each symbol, which requires serial processing that prevents efficient GPU / TPU utilization. We introduce two enhancements, channel-conditioning and latent residual prediction, that lead to network architectures with better rate-distortion performance than existing context-adaptive models while minimizing serial processing. Empirically, we see an average rate savings of 6.7% on the Kodak image set and 11.4% on the Tecnick image set compared to a context-adaptive baseline model. At low bit rates, where the improvements are most effective, our model saves up to 18% over the baseline and outperforms hand-engineered codecs like BPG by up to 25%.

Channel-Wise Autoregressive Entropy Models for Learned Image Compression