Practical Binary Adaptive Block Coder

TL;DR

This paper presents a low-complexity adaptive binary encoding algorithm using universal block codes based on context, with theoretical redundancy analysis and experimental comparison to existing methods.

Contribution

It introduces a novel adaptive encoding algorithm with a refined redundancy formula and demonstrates its effectiveness against established encoders.

Findings

The redundancy formula improves previous estimates.

The algorithm achieves competitive performance in experiments.

Experimental results compare favorably with JBIG and MPEG encoders.

Abstract

This paper describes design of a low-complexity algorithm for adaptive encoding/ decoding of binary sequences produced by memoryless sources. The algorithm implements universal block codes constructed for a set of contexts identified by the numbers of non-zero bits in previous bits in a sequence. We derive a precise formula for asymptotic redundancy of such codes, which refines previous well-known estimate by Krichevsky and Trofimov, and provide experimental verification of this result. In our experimental study we also compare our implementation with existing binary adaptive encoders, such as JBIG's Q-coder, and MPEG AVC (ITU-T H.264)'s CABAC algorithms.