Long-term background memory based on Gaussian mixture model
Wenbo Zhao, Xudong Zhao, W.M. Liu, Xianfeng Tang · 2013
This paper aims to present a long-term background memory framework, which is capable of memorizing long period background in video and rapidly adapting to the changes of background. Based on Gaussian mixture model (GMM), this framework enables an accurate identification of long period background appearances and presents a perfect solution to numerous typical problems on foreground detection. The experimental results with various benchmark sequences quantitatively and qualitatively demonstrate that the proposed algorithm outperforms many GMM-based methods for foreground detection, as well as other representative approaches.