Intelligent Engineering Systems through Artificial Neural Networks, Volume 16
49 Analysis of Human Gait Using Motion Capture Data and Self-Organizing Map Paradigms
Download citation file:
- Ris (Zotero)
- Reference Manager
Studies of human perception show that human observers can almost instantaneously recognize biological motion patterns. These patterns can be decomposed into identifiable elements, which we organize in a hierarchical structure. The objective of this research is to find such a motion hierarchy using the visual information collected through motion capture and the learning algorithms of Self-organizing Maps (SOM). Such structure is sought to provide a machine with the ability to recognize biological motion patterns and to control the generality, precision, and realism in generating motion models for clinical studies, animation, and other disciplines.