Multi-class moving target detection with Gaussian mixture part based model

This paper proposes an effective multi-class moving target detection system that is based on Gaussian-mixture part-based model (GM-PBM), which accurately locates objects of interest and recognizes their corresponding category. This system is multi-threaded and combines soft clustering approach with multiple mixture part-based models to provide stable multi-class target tracking and recognition in videos. Experimental results show that real-time simultaneous detection and tracking of multi-class objects is viable using the mentioned system.