作者: Yi Zhang , Jie Yang , Kun Liu
关键词:
摘要: Effective and robust recognition tracking of objects are the key problems in visual surveillance systems. Most existing object methods were designed with particular mind. This study presents a general moving method using global features targets. Targets extracted an adaptive Gaussian mixture model their silhouette images captured unified. A new database is built to provide abundant samples train subspace feature. more convincing than previous ones. effective dimension reduction based on graph embedding used obtain projection eigenvector. In our experiments, we show performance addressing problem its superiority compared methods.