RECOMBINER: Robust and Enhanced Compression with Bayesian Implicit Neural Representations

TL;DR

RECOMBINER advances data compression by enhancing Bayesian implicit neural representations with flexible priors, local detail adaptation, and patch-based robustness, outperforming previous INR methods and autoencoders at low bitrates.

Contribution

It introduces RECOMBINER, a novel INR-based compression method that improves flexibility, local adaptation, and robustness over prior approaches.

Findings

Achieves competitive results with top INR-based methods.

Outperforms autoencoder codecs on low-resolution images at low bitrates.

Demonstrates effectiveness across multiple data modalities.

Abstract

COMpression with Bayesian Implicit NEural Representations (COMBINER) is a recent data compression method that addresses a key inefficiency of previous Implicit Neural Representation (INR)-based approaches: it avoids quantization and enables direct optimization of the rate-distortion performance. However, COMBINER still has significant limitations: 1) it uses factorized priors and posterior approximations that lack flexibility; 2) it cannot effectively adapt to local deviations from global patterns in the data; and 3) its performance can be susceptible to modeling choices and the variational parameters' initializations. Our proposed method, Robust and Enhanced COMBINER (RECOMBINER), addresses these issues by 1) enriching the variational approximation while retaining a low computational cost via a linear reparameterization of the INR weights, 2) augmenting our INRs with learnable positional encodings that enable them to adapt to local details and 3) splitting high-resolution data into patches to increase robustness and utilizing expressive hierarchical priors to capture dependency across patches. We conduct extensive experiments across several data modalities, showcasing that RECOMBINER achieves competitive results with the best INR-based methods and even outperforms autoencoder-based codecs on low-resolution images at low bitrates. Our PyTorch implementation is available at https://github.com/cambridge-mlg/RECOMBINER/.