Automatic Video Abstraction via the Progress of Story
In this paper, an automatic video abstraction scheme on continuously recorded video, such as a movie video, is proposed. Conventional methods deal with the issue of video abstraction from the level of scene, while the proposed method attempts to comprehend video contents from the progress of the overall story and viewers’ semantic understanding. The generated dynamic abstraction not only provides a bird view of the original video but also helps a viewer understand the progress of the overall story. Furthermore, different types of video abstraction can be appropriately generated with respect to different user-defined duration length. Experimental results show that the proposed scheme is a feasible solution for the effective management of video repository and online review services.
KeywordsAutomatic video abstraction progress of a story semantic understanding
Unable to display preview. Download preview PDF.
- 2.Ma, Y., Lu, L., Zhang, H., Li, M.: A user attention model for video summarization. In: ACM International Conference on Multimedia, pp. 533–542. ACM Press, Juan les Pins France (2002)Google Scholar
- 3.Ma, Y., Lu, L., Zhang, H.: Video Snapshot: A bird view of video sequence. In: IEEE Conference on Multimedia Modeling, pp. 94–101. IEEE Press, Melbourne (2005)Google Scholar
- 8.Zhu, S., Liu, Y.: A Novel Scheme for Video Scenes Segmentation and Semantic Representation. In: IEEE Conference on Multimedia and Expo., pp. 1289–1292. IEEE Press, Hannover (2008)Google Scholar
- 9.Li, S., Zhu, L., Zhang, Z., Blake, A., Zhang, H., Shum, H.: Statistic learning of multi-view face detection. In: Europe Conference on Computer Vision, pp. 117–121. Europe Press, Copenhagen (2002)Google Scholar