Dictionary-based map compression using geometric priors

Obtaining a compact representation of a given pointset map built by mapper robots is a critical issue for recent SLAM applications. This “map compression” problem is explored from a novel viewpoint of dictionary-based map compression techniques in the paper. The primary contribution of the paper is on the use of geometric priors within a dictionary-based map compression framework. An efficient map compressor is presented using RANSAC map-matching as well as a recursive map matching scheme. The presented techniques are experimentally evaluated in terms of compression ratio as well as compression speed using radish dataset.