A Video Retrieval Algorithm Based on Affective Features

One important but often overlooked aspect of human interpretation of multimedia data is the affective information. Affective labels of video content can be extracted automatically from multimedia data streams. These can then be used for content-based retrieval and browsing. In this paper we propose an approach to retrieve the video clips which are similar to the given query in emotion from the video database. We adopt fuzzy representation of video emotion to bridge the semantic gap. Then, we use the cosine measure to compute the similarity between the query video clip and each video clip in database. Ten video clips in database with the highest similarity are output and submitted to the user. And our experiment indicates that the algorithm works well.