Efficient Indexing of 3D Human Motions

Logo poskytovatele

Varování

Publikace nespadá pod Fakultu sportovních studií, ale pod Fakultu informatiky. Oficiální stránka publikace je na webu muni.cz.
Autoři

BUDÍKOVÁ Petra SEDMIDUBSKÝ Jan ZEZULA Pavel

Rok publikování 2021
Druh Článek ve sborníku
Konference ACM International Conference on Multimedia Retrieval (ICMR)
Fakulta / Pracoviště MU

Fakulta informatiky

Citace
www https://dl.acm.org/doi/10.1145/3460426.3463646
Doi http://dx.doi.org/10.1145/3460426.3463646
Klíčová slova human motion data; skeleton sequences; motion word; text-based processing; indexing; extended inverted files; ranked retrieval; approximate searching; scalability
Popis Digitization of human motion using 2D or 3D skeleton representations offers exciting possibilities for many applications but, at the same time, requires scalable content-based retrieval techniques to make such data reusable. Although a lot of research effort focuses on extracting content-preserving motion features, there is a lack of techniques that support efficient similarity search on a large scale. In this paper, we introduce a new indexing scheme for organizing large collections of spatio-temporal skeleton sequences. Specifically, we apply the motion-word concept to transform skeleton sequences into structured text-like motion documents, and index such documents using an extended inverted-file approach. Over this index, we design a new similarity search algorithm that exploits the properties of the motion-word representation and provides efficient retrieval with a variable level of approximation, possibly reaching constant search costs disregarding the collection size. Experimental results confirm the usefulness of the proposed approach.
Související projekty:

Používáte starou verzi internetového prohlížeče. Doporučujeme aktualizovat Váš prohlížeč na nejnovější verzi.

Další info