Abstract
This paper describes a novel approach for incremental symbolization and classification of observed human motion. The acquired symbols can be used for both subsequent motion recognition and generation. The resulting representation is a tree structure with the symbol at each node corresponding to its abstraction level.