End-to-end optimized image compression with competition of prior distributions
This work addresses the computational bottleneck in entropy coding for image compression, offering a more efficient solution for applications requiring low-latency or resource-constrained environments, though it is incremental as it builds on existing autoencoder-based methods.
The paper tackles the problem of reducing complexity in learned image compression by replacing a second autoencoder for prior prediction with a static table of multiple learned priors selected via competition, achieving comparable rate-distortion performance with significantly lower entropy coding and decoding complexity.
Convolutional autoencoders are now at the forefront of image compression research. To improve their entropy coding, encoder output is typically analyzed with a second autoencoder to generate per-variable parametrized prior probability distributions. We instead propose a compression scheme that uses a single convolutional autoencoder and multiple learned prior distributions working as a competition of experts. Trained prior distributions are stored in a static table of cumulative distribution functions. During inference, this table is used by an entropy coder as a look-up-table to determine the best prior for each spatial location. Our method offers rate-distortion performance comparable to that obtained with a predicted parametrized prior with only a fraction of its entropy coding and decoding complexity.