25 October 2016 Recognition of complex human behaviours using 3D imaging for intelligent surveillance applications
Author Affiliations +
Abstract
We introduce a system that exploits 3-D imaging technology as an enabler for the robust recognition of the human form. We combine this with pose and feature recognition capabilities from which we can recognise high-level human behaviours. We propose a hierarchical methodology for the recognition of complex human behaviours, based on the identification of a set of atomic behaviours, individual and sequential poses (e.g. standing, sitting, walking, drinking and eating) that provides a framework from which we adopt time-based machine learning techniques to recognise complex behaviour patterns.
Conference Presentation
© (2016) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Bo Yao, Bo Yao, Jason J. Lepley, Jason J. Lepley, Robert Peall, Robert Peall, Michael Butler, Michael Butler, Hani Hagras, Hani Hagras, } "Recognition of complex human behaviours using 3D imaging for intelligent surveillance applications", Proc. SPIE 9992, Emerging Imaging and Sensing Technologies, 99920H (25 October 2016); doi: 10.1117/12.2241088; https://doi.org/10.1117/12.2241088
PROCEEDINGS
5 PAGES + PRESENTATION

SHARE
Back to Top