Object tracking in compressed MPEG video bitstream.

摘要

Given that video is stored efficiently in a compressed format, the costly overhead of decompression can be reduced by analyzing the compressed representation directly. Performing analysis in the compressed domain reduces the amount the data, which keeps the storage costs of the resulting feature set low.; Moving Picture Experts Group (MPEG) has developed the following generic audio-video standards to date: MPEG-1, MPEG-2, MPEG-4 MPEG-7 and MPEG-21. MPEG video compression is used in many current and emerging products. It is at the heart of digital television set-top boxes, DSS, HDTV decoders, DVD players, video conferencing, Internet video, and other applications. These applications benefit from video compression in the fact that they may require less storage space for archived video information, less bandwidth for the transmission of the video information from one point to another, or a combination of both. While MPEG-1 was developed for storage and retrieval of moving pictures and audio on storage media, MPEG-2 targeted higher quality and bandwidth applications like digital television.; This thesis evaluates the efficacy of using MPEG encoded video as a means for more efficient object tracking. Real-time MPEG encoding of video is available in hardware and the compressed data stream produced contains tracking information as a by-product of the encoding process. Our interest here is to quantify and extract that data in real-time and determine if the procedure represents a viable approach to object tracking against existing techniques.