Parametric context adaptive Laplace distribution for multimedia compression

TL;DR

This paper introduces a new adaptive Laplace distribution model for multimedia data compression that leverages autoregressive context-dependent parameters, improving efficiency and generalization over traditional quantization methods.

Contribution

It proposes inexpensive autoregressive models for context-dependent Laplace distribution parameters, enabling better adaptation and higher-dimensional context utilization in multimedia compression.

Findings

Achieved similar performance to LOCO-I with fewer parameters.

Reduced header size and improved generalization.

Enabled use of higher-dimensional contexts like multiple color channels.

Abstract

Data compression often subtracts prediction and encodes the difference (residue) e.g. assuming Laplace distribution, for example for images, videos, audio, or numerical data. Its performance is strongly dependent on the proper choice of width (scale parameter) of this parametric distribution, can be improved if optimizing it based on local situation like context. For example in popular LOCO-I \cite{loco} (JPEG-LS) lossless image compressor there is used 3 dimensional context quantized into 365 discrete possibilities treated independently. This article discusses inexpensive approaches for exploiting their dependencies with autoregressive ARCH-like context dependent models for parameters of parametric distribution for residue, also evolving in time for adaptive case. For example tested such 4 or 11 parameter models turned out to provide similar performance as 365 parameter LOCO-I model for 48 tested images. Beside smaller headers, such reduction of number of parameters can lead to better generalization. In contrast to context quantization approaches, parameterized models also allow to directly use higher dimensional contexts, for example using information from all 3 color channels, further pixels, some additional region classifiers, or from interleaving multi-scale scanning - for which there is proposed Haar upscale scan combining advantages of Haar wavelets with possibility of scanning exploiting local contexts.