News video abstract extraction based on corner detection

Video abstraction plays an important role for news video and is one of the main techniques for video retrieval and skimming. Consequently, it is extremely important that how to extract the key-frames, which reflect the content of the video exactly, to set up the video abstraction. To solve such a problem, this paper proposes a new method, which uses the technique of corner detection to identify the subject caption frames in gray-scale color space. Then the video abstraction is set up by extracting reasonable caption frames under certain regulations. Experiment results show the efficiency of the proposed algorithm for news video abstraction. Copyright © 2010 Binary Information Press.