Hybrid approach for event detection from video surveillance sequences

In this paper, we propose a strategy of combined SVMs learning system for detection of predefined events in the video. The extraction and synthesis of suitably visual descriptor based on moving object traking are used for this purpose. The results showed that the performance of this combined classifiers is improving vs a single SVM. We then demonstrate the usefulness of the toolbox in the context of feature extraction, events learning and detection in large collection of video surveillance dataset.