Correlation as a measure for alignment and similarity of human motions

Randall, Mathew and Harvey, Carlo and Williams, Ian (2023) Correlation as a measure for alignment and similarity of human motions. Computer Animation and Virtual Worlds, 34 (3-4). e2157. ISSN 1546-4261

Computer Animation Virtual - 2023 - Randall - Correlation as a measure for alignment and similarity of human motions.pdf - Published Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (2MB)


The ability to measure similarity and alignment of motions is a key tool in motion retrieval and motion editing. Similarity metrics based on distance functions are often utilized when measuring similarity of human motions, however, metrics based on correlation can also potentially useful for measuring similarity and alignment. This paper evaluates the use of correlation as a method of measuring the alignment and similarity of human motion and compares them against more established distance-based metrics. Three correlation methods and five methods of parameterising rotation are evaluated. The results show that parameterization based on displacement vectors and Kendall Tau rank correlation are optimal for measuring the alignment between two motions. If measuring similarity of motions, however, an approach based on distance metrics for angular or positional distance should be used.

Item Type: Article
Identification Number:
1 May 2023Accepted
16 May 2023Published Online
Uncontrolled Keywords: human motion, motion similarity, motion alignment
Subjects: CAH03 - biological and sport sciences > CAH03-02 - sport and exercise sciences > CAH03-02-01 - sport and exercise sciences
Divisions: Faculty of Computing, Engineering and the Built Environment > School of Computing and Digital Technology
Depositing User: Gemma Tonks
Date Deposited: 08 Jun 2023 10:44
Last Modified: 08 Jun 2023 10:44

Actions (login required)

View Item View Item


In this section...