Search

Highlights: [•] An algorithm that parses 3D-cloud representing the human body into torso and arms. [•] A technique to locate the held object, and incorporate size into a recognizer. [•] A temporal smoothing scheme to improve object and activity recognition. [Copyright &y& Elsevier]

Highlights: [•] A new trilateral filter incorporating distance, color, and boundary information. [•] A new Graph Cuts based on color, depth, and boundary information. [•] A method to rectify the misaligned depth and RGB data of fast moving objects. [Copyright &y& Elsevier]

Abstract: Most automatic event detection methods for video surveillance detect target events based on features extracted in the pixel domain. However, in practice, surveillance videos are often compressed. It is desirable to perform automatic event detection in the compressed domain directly so that the video does not need to be decoded for analysis purpose. In this paper, we investigate the use of motion trajectories for video activity detection in the compressed domain. We show it is possible to extract reliable motion trajectories directly from compressed H.264 video streams. To overcome the problems caused by unreliable motion vectors, we propose to include the information from the compressed domain prediction residuals to make the tracking more robust. We use a real world application of detecting vacant or occupied parking spaces to demonstrate the effectiveness of our proposed approach. We also demonstrate the robustness of our approach to different encoder settings, and lighting conditions. [Copyright &y& Elsevier]