Human pose recognition by Memory-Based Hierarchical Feature Matching

摘要

A human posture recognition system based on a memory-based approach is studied. Human body images extracted by depth data are labeled and stored in a database together with compressed feature values, i.e., higher-order local correlation values and outline diameters. The compressed features are used to speed up the database search in a hierarchical manner. In the experiment, 8 postures are measured and classified with Primaly Component Analysis. This system can recognize the postures using memory-based data matching under the classification.