欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
提出一种基于时空兴趣点的人体行为识别算法,该算法从行为视频中提取丰富的时空兴趣点,然后在数据空间采用基于高斯混合模型的聚类方法完成对特征点的分类.在动作识别过程中,采用平均Hausdorff距离来衡量序列间匹配的相似性,提高运算效率.KTH数据库上的实验证明了该算法的有效性和鲁棒性.
英文摘要：
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 KTH database prove the effectiveness and robustness of the algorithm.
同期刊论文项目
同项目期刊论文
期刊信息
位置：