Title :
Scene content classification from MPEG coded bit streams
Author :
Dawood, Ali M. ; Ghanbari, Mohammed
Author_Institution :
Dept. of Electron. Syst. Eng., Essex Univ., Colchester, UK
Abstract :
We propose a classification method of MPEG coded video based on the image texture and motion levels of the scene content. The texture and motion are divided into three levels each, low, medium and high, resulting into nine scene categories. Since in MPEG-1 (and MPEG-2) encoder the DCT coefficients of the I-frame are affected by the image texture, while the motion affects the macroblocks coding mode of P- and B-frames, hence, an automatic scene classification can be obtained from the coded bit stream based on the scene´s texture and motion levels. The method is useful for the content-based video analysis, such as video modeling and MPEG-7
Keywords :
code standards; data compression; discrete cosine transforms; image classification; image motion analysis; image texture; telecommunication standards; transform coding; video coding; B-frames; DCT coefficients; I-frame; MPEG coded bit streams; MPEG coded video; MPEG-1 encoder; MPEG-2 encoder; MPEG-7 standard; P-frames; automatic scene classification; content-based video analysis; image texture; macroblocks coding mode; motion levels; scene content classification; video modeling; Data mining; Discrete cosine transforms; Image coding; Image databases; Image texture; Layout; MPEG 7 Standard; Multimedia databases; Streaming media; Videoconference;
Conference_Titel :
Multimedia Signal Processing, 1999 IEEE 3rd Workshop on
Conference_Location :
Copenhagen
Print_ISBN :
0-7803-5610-1
DOI :
10.1109/MMSP.1999.793841
