Reducing False Positives in Video Shot Detection Using Learning Techniques
Video has become an interactive medium of daily use today. However, the sheer volume of the data makes it extremely difficult to browse and find required information. Organizing the video and locating required information effectively and efficiently presents a great challenge to the video retrieval community. This demands a tool which would break down the video into smaller and manageable units called shots.
Traditional shot detection methods use pixel difference, histograms, or temporal slice analysis to detect hard-cuts and gradual transitions. However, systems need to be robust to sequences that contain dramatic illumination changes, shaky camera effects, and special effects such as fire, explosion, and synthetic screen split manipulations. Traditional systems produce false positives for these cases; i.e., they claim a shot break when there is none.
We propose a shot detection system which reduces false positives even if all the above effects are cumulatively present in one sequence. Similarities between successive frames are computed by finding the correlation and is further analyzed using a wavelet transformation. A final filtering step is to use a trained Support Vector Machine (SVM). As a result, we achieve better accuracy (while retaining speed) in detecting shot-breaks when compared with other techniques.
KeywordsFalse Positive Gradual Transition Illumination Change Morlet Wavelet Successive Frame
Unable to display preview. Download preview PDF.
- 2.Bang, C., Chenl, S.C., Shyu, M.L.: Pixso: a system for video shot detection. In: Fourth International Conference on Information, Communications and Signal Processing, pp. 1320–1324 (2003)Google Scholar
- 3.Shahraray, S.: Scene change detection and content-based sampling of video sequence. In: SPIE Storage and Retrieval for Image and Video Databases, pp. 2–13 (1995)Google Scholar
- 4.Swanberg, D., Shu, C., Jain, R.: Knowledge guided parsing in video database. In: SPIE Storage and Retrieval for Image and Video Databases, pp. 13–24 (1993)Google Scholar
- 6.Rasheed, Z., Shah, M.: Scene detection in Hollywood movies and TV shows. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 343–348 (2003)Google Scholar
- 8.Li, D., Lu, H.: Avoiding false alarms due to illumination variation in shot detection. In: IEEE Workshop on Signal Processing Systems, pp. 828–836 (2000)Google Scholar
- 9.Lu, H., Tan, Y.: An effective post-refinement method for shot boundary detection. CirSysVideo 15, 1407–1421 (2005)Google Scholar
- 10.Zabih, R., Miller, J., Mai, K.: Feature-based algorithms for detecting and classifying scene breaks. Technical report, Cornell University (1995)Google Scholar
- 11.Yuliang, G., De, X.: A solution to illumination variation problem in shot detection. In: TENCON 2004. IEEE Region 10 Conference, pp. 81–84 (2004)Google Scholar
- 12.Ngo, C., Pong, T., Chin, R.: Detection of gradual transitions through temporal slice analysis. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 36–41 (1999)Google Scholar
- 13.Yeo, C., Zhu, Y.W., Sun, Q., Chang, S.F.: A framework for sub-window shot detection. In: MMM 2005: Eleventh International Multimedia Modelling Conference (MMM 2005), pp. 84–91 (2005)Google Scholar
- 14.Amir, A., et al.: IBM Research TRECVID-2005 Video Retrieval System. In: TREC Proc. (2005)Google Scholar
- 17.Petersohn, C.: Fraunhofer HHI at TRECVID 2004: Shot boundary detection system. In: TREC Proc. (2004)Google Scholar
- 18.Covell, M., Ahmad, S.: Analysis by synthesis dissolve detection. In: International Conference on Image Processing, pp. 425–428 (2002)Google Scholar
- 19.Lienhart, R., Zaccarin, A.: A system for reliable dissolve detection in videos. In: International Conference on Image Processing, pp. 406–409 (2001)Google Scholar
- 20.NIST: TREC Video Retrieval Evaluation (2005), http://www-nlpir.nist.gov/projects/trecvid