Human action analysis and recognition are increasingly attracting much attention from computer vision and pattern recognition researchers.This paper presents a human action recognition based on space-time interest point algorithm.The algorithm detects dense space-time interest points from action videos and classifies these feature points using cluster method based on Gaussian mixture model in the data space.In the process of action recognition

a matching-based approach with the mean Hausdorff distance is measured the similarity between image sequences to improve the operation efficiency.The experiments on the KT H database prove the effectiveness and robustness of the algorithm.