Tutorial 1: Similarity Search in 3D Human Motion Data

Please check the pdf version of the tutorial

Motion capture technologies can digitize human movements into a discrete sequence of 3D skeletons. Such spatio-temporal data have a great application potential in many fields, ranging from computer animation, through security and sports to medicine, but their computerized processing is a difficult problem.

The objective of this tutorial is to explain fundamental principles and technologies designed for searching, subsequence matching, classification and action detection in the 3D human motion data. These operations inherently require the concept of similarity to determine the degree of accordance between pairs of 3D skeleton sequences. Such similarity can be modeled using a generic approach of metric space by extracting effective deep features and comparing them by efficient distance functions. The metric-space approach also enables applying traditional index structures to efficiently access large datasets of skeleton sequences. We demonstrate the functionality of selected motion-processing operations by interactive web applications.


Jan Sedmidubsky

Jan Sedmidubsky is a researcher of computer science at Masaryk University (Czech Republic) where he received the Ph.D. degree in 2011 along with the dean’s and rector’s prize for a distinguished dissertation thesis. His research activities are primarily concentrated on similarity processing of 3D human motion data. He is a co-author of about 40 research publications.

Pavel Zezula

Pavel Zezula is a professor of computer science at Masaryk University (Czech Republic). His professional interests primarily concern multimedia content-based retrieval, large-scale similarity search, and big data analysis. He is a co-author of seminal similarity search structure, the "M-Tree", and the book“Similarity Search: The Metric Space Approach” by Springer US. He is also a co-author of more than 150 research publications with more than 6,000 citations.