Automatic categorization and topic tracking techniques for Web news videos
Hua Huo · Journal of Hefei University of Technology · 2012
Automatic categorization and topic tracking of Web news videos are a challenging task.Critical semantic elements are extracted for news videos analysis.An automatic news video classification method based on hierarchical conditional random field is proposed to fuse multimode features of text and vision.Then an incremental k-means clustering method is designed to track the videos with the same topic.Extensive experiments on seven most frequent categories in Youku Website show the effectiveness of the proposed method for categorizing and tracking Web news videos.