PCA-based hierarchical approach for video gradual transition detection in compressed domain

A novel automatic real-time video parsing framework for gradual transition based on principal component analysis is presented. In this framework, three core factors of video parsing that is feature space, discontinuity metric and decision rule are stressed to improve the parsing performance. Through this PCA-based framework, the negative influence of significant camera or objects movement to the gradual transition can be greatly depressed, and thus the performance of video parsing is greatly improved. This framework is implemented on the TREC video test repository to validate its performance.