SEDMIDUBSKÝ, Jan, Petra BUDÍKOVÁ, Vlastislav DOHNAL and Pavel ZEZULA. Motion Words: A Text-like Representation of 3D Skeleton Sequences. Online. In 42nd European Conference on Information Retrieval (ECIR). Cham: Springer, 2020, p. 527-541. ISBN 978-3-030-45438-8. Available from: https://dx.doi.org/10.1007/978-3-030-45439-5_35.
Other formats:   BibTeX LaTeX RIS
Basic information
Original name Motion Words: A Text-like Representation of 3D Skeleton Sequences
Authors SEDMIDUBSKÝ, Jan (203 Czech Republic, guarantor, belonging to the institution), Petra BUDÍKOVÁ (203 Czech Republic, belonging to the institution), Vlastislav DOHNAL (203 Czech Republic, belonging to the institution) and Pavel ZEZULA (203 Czech Republic, belonging to the institution).
Edition Cham, 42nd European Conference on Information Retrieval (ECIR), p. 527-541, 15 pp. 2020.
Publisher Springer
Other information
Original language English
Type of outcome Proceedings paper
Field of Study 10200 1.2 Computer and information sciences
Confidentiality degree is not subject to a state or trade secret
Publication form electronic version available online
Impact factor Impact factor: 0.402 in 2005
RIV identification code RIV/00216224:14330/20:00114026
Organization unit Faculty of Informatics
ISBN 978-3-030-45438-8
ISSN 0302-9743
Doi http://dx.doi.org/10.1007/978-3-030-45439-5_35
Keywords in English 3D skeleton sequence;motion word;motion vocabulary;quantization;border problem;text-based processing
Tags core_A, DISA, firank_A
Tags International impact, Reviewed
Changed by Changed by: RNDr. Pavel Šmerk, Ph.D., učo 3880. Changed: 29/4/2021 12:21.
Abstract
There is a growing amount of human motion data captured as a continuous 3D skeleton sequence without any information about its semantic partitioning. To make such unsegmented and unlabeled data efficiently accessible, we propose to transform them into a text-like representation and employ well-known text retrieval models. Specifically, we partition each motion synthetically into a sequence of short segments and quantize the segments into motion words, i.e. compact features with similar characteristics as words in text documents. We introduce several quantization techniques for building motion-word vocabularies and propose application-independent criteria for assessing the vocabulary quality. We verify these criteria on two real-life application scenarios.
Links
GA19-02033S, research and development projectName: Vyhledávání, analytika a anotace datových toků lidských pohybů
Investor: Czech Science Foundation
PrintDisplayed: 25/4/2024 15:40