Stochastic search algorithms for optimal content-based sampling of video sequences

A.D. Doulamis1, N.D. Doulamis1
1Department of Electrical and Computer Engineering, National and Technical University of Athens, Zografou, Greece

Tóm tắt

A video content representation framework is proposed in this paper, for extracting limited, but meaningful, information of video data, directly from the MPEG compressed domain. In particular, extraction of several representative shots is performed for each video sequence in a content based rate sampling framework. An approach, based on minimization of a cross-correlation criterion of the video frames has been adopted for the shot selection. For efficient implementation of the latter approach, a logarithmic search in a stochastic framework is proposed. The method always converges to the global minimum as is proven in the paper.

Từ khóa

#Stochastic processes #Sampling methods #Video sequences #Video on demand #Multimedia systems #Data mining #Videoconference #Communications technology #Information retrieval #Computer aided instruction

Tài liệu tham khảo

10.1109/CVPR.1998.698631 10.1109/76.844996 10.1109/30.883444 10.1109/MMCS.1999.778618 10.1006/rtim.1999.0185 10.1109/76.736718 10.1109/76.633496 riu, 0, Relevance Feedback: A Power Tool for Interactive Content-based Image Retrieval, IEEE Trans Cir Sys Video Tech, 8, 644 10.1109/5.664279 10.1006/cviu.1999.0761 10.1109/76.633489 10.1109/76.718508 yeo, 1995, Rapid Scene Analysis on Compressed Videos, IEEE Trans Circuits and Systems for Video Technology, 5, 533, 10.1109/76.475896