A Survey of Video Parsing and Image Indexing Techniques in Compressed Domain

Compressed Domain C. W. Ngo, T. C. Pong & R. T. Chin Department of Computer Science The Hong Kong University of Science & Technology Clear Water Bay, Kowloon, Hong Kong fcwngo, tcpong, [email protected] Abstract| With the wide spread use of WWW, digital libraries are expected to handle more visual information in the future. Furthermore, most of the data are represented in compressed formats due to the limitations of time and space. As a result, most techniques have been developed to index visual features directly in the compressed domain. These techniques omit the decompression process and, in addition, reduce processing time since relatively small amount of data are analyzed. In this paper, we mainly review recent parsing and indexing techniques operated in the compressed domain. Related issues include image feature extraction, video segmentation, and video modeling.