Adaptive Background Modeling with Temporal Feature Update for Dynamic Foreground Object Removal

This thesis explores several approaches in order to develop an effective algorithm for separating the moving objects from the background so that the background alone may be displayed in a video sequence. In particular, this thesis seeks to develop an algorithm that can successfully remove a moving teacher from the whiteboard, so that the writing on the whiteboard may be fully visible to the audience. In this thesis, some prior works related to this problem are studied to understand the basic principles of adaptive background modeling, or removing the foreground from a background with occasional changes in a video sequence. Three different algorithms are developed for this purpose. The design and implementation of these algorithms are discussed in detail, and the results are presented. Lastly, the results of this research are generalized and potential future research is discussed