Flexible multi-objective particle swarm optimization clustering with game theory to address human activity discovery fully unsupervised

• Published in: Image and vision computing Vol. 145; p. 104985
• Main Authors: Hadikhani, Parham, Lai, Daphne Teck Ching, Ong, Wee-Hong
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.05.2024
• Subjects: Clustering; Dimension reduction; Feature extraction; Human activity discovery; Incremental manner; Multi-objectives optimization; Skeleton sequence; Unsupervised learning; Dimension reduction; Incremental manner; Skeleton sequence; Human activity discovery; Multi-objectives optimization; Feature extraction; Clustering; Unsupervised learning
• ISSN: 0262-8856; 1872-8138
• DOI: 10.1016/j.imavis.2024.104985

Human activity recognition is a crucial field of study, but current approaches often require ground truth labels, which are not always available. We propose a new method called the Flexible Multi-Objective Particle swarm optimization clustering method based on Game theory (FMOPG), which can identify human activities without any supervision. Unlike traditional clustering methods that require an estimate of the number of clusters and are often inaccurate, FMOPG handles varying cluster numbers with an incremental technique, selecting clusters with good connectivity and separation. We enhance Particle Swarm Optimization (PSO) with mean-shift vectors for faster convergence and better handling of non-spherical clusters. Employing multi-objective optimization and Gaussian mutation, FMOPG provides a range of optimal solutions. We map the optimization problem to game theory to select the best solution based on different criteria. A smart grid-based method is proposed for population initialization, reducing variance and improving reliability. FMOPG outperforms state-of-the-art methods, improving clustering accuracy by 3.65%. Moreover, the incremental technique has improved clustering time by 71.18%. •Novel unsupervised method for discovering human activity from skeleton-based data.•Introducing a flexible multi-objective PSO clustering based on game theory.•Adopting Incremental techniques to estimate the number of activities automatically.•Proposing smart grid-based swarm initialization to generate diverse solutions.•Updating particle velocity based on mean shift to find nonlinear clusters.