Title :
Inertia-based cut detection technique: a step to the integration of video coding and content-based retrieval
Author :
Liu, Tieyun ; Xudong Zhang ; Desheng Wang ; Feng, Jian ; Lo, Kwok-Tung
Author_Institution :
Dept. of Electron. Eng., Tsinghua Univ., Beijing, China
Abstract :
A novel cut detection algorithm as well as a related novel coding scheme is developed to serve as a step to the integration of video coding and content-based retrieval. The cut detection algorithm consists of inertia-based motion pre-compensation, modified block-based histogram differencing, adaptive thresholding and fade normalization. By use of these techniques, a frame can be judged to be a cut or not accurately before normal motion estimation and compensation. At the same time, under the framework of the new coding scheme useful information about shot boundary and key frames can be generated and embedded in the bit stream so that people can browse, index and match video shots more conveniently
Keywords :
content-based retrieval; image retrieval; motion compensation; motion estimation; video coding; adaptive thresholding; content-based retrieval; fade normalization; inertia-based cut detection technique; inertia-based motion pre-compensation; key frames; modified block-based histogram differencing; motion compensation; motion estimation; shot boundary; video coding; video shots indexing; video shots matching; Content based retrieval; Detection algorithms; Histograms; Motion compensation; Motion detection; Motion estimation; Signal processing algorithms; Streaming media; Video coding; Video sequences;
Conference_Titel :
Signal Processing Proceedings, 2000. WCCC-ICSP 2000. 5th International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-5747-7
DOI :
10.1109/ICOSP.2000.891700