Video is a powerful tool to show various activities but generally we use still images to show a condensed video, which is problematic in viewing and comprehending. Thus, there is a need for a summarized surveillance video. A fundamental goal of any video summarization or synopsis technique with reference to a surveillance video is to reduce the Spatio-temporal redundancy. The activity in any surveillance video is very less as compared to the total length of the video. The spatial redundancy is removed by showing two activities that happened in different frames at different spatial locations in a single frame. Temporal redundancy is removed by detecting the frames having low activity and then deleting those frames. We then generate a stroboscopic video, which traces path of the extracted object. Lastly, we introduce a Media Player, which indexes the video synopsis to the original video demonstrating how the video synopsis can be used as an effective tool.
展开▼
