Comparison of content selection methods for skimming rushes video

ABSTRACT

We compare two methods for selecting segments to be included in a video skim, using lists of relevant as well as redundant segments created from different visual features as input. One approach is rule-based, and creates a weighted sum of the input relevances. The other is HMM based, using a model trained on the TRECVID 2007 rushes data. The redundant segments are created from detection of repeated takes and junk content, the selected segments from visual activity and face detection. The results show that the approaches create very short summaries which only contain a part of the relevant information in the video, but reach very high scores in terms of the usability measures non-duplicates, non-junk and pleasant tempo. The HMM based approach contains more information despite shorter duration of the summaries.

REFERENCES

• 1: W. Bailer. A comparison of distance measures for clustering video sequences. In Proceedings of 1st Workshop on Automated Information Extraction in Media Production, Turin, IT, Sept. 2008.
• 2: W. Bailer, E. Dumont, S. Essid, and B. Mérialdo. A collaborative approach to automatic rushes video summarization. In Proceedings of IEEE International Conference on Image Processing, San Diego, CA, USA, Sept. 2008.
• 3: Werner Bailer , Felix Lee , Georg Thallinger, Skimming rushes video using retake detection, Proceedings of the international workshop on TRECVID video summarization, p.60-64, September 28-28, 2007, Augsburg, Bavaria, Germany [doi>10.1145/1290031.1290042]
• 4: W. Bailer, F. Lee, and G. Thallinger. Detecting and clustering multiple takes of one scene. In MMM, pages 80--89, Kyoto, JP, Jan. 2008.
• 5: C.-C. Chang and C.-J. Lin. LIBSVM: a library for support vector machines, 2001.
• 6: MPEG-7. Multimedia Content Description Interface-Part 3: Visual. ISO/IEC 15938--3, 2001.
• 7: NHK STRL. Test modules for TRECVID activity. Use case scenario. Ver.1.2.0E, Apr. 2008.
• 8: Paul Over , Alan F. Smeaton , George Awad, The trecvid 2008 BBC rushes summarization evaluation, Proceedings of the 2nd ACM TRECVid Video Summarization Workshop, p.1-20, October 31-31, 2008, Vancouver, British Columbia, Canada [doi>10.1145/1463563.1463564]
• 9: A. F. Smeaton and P. Over. TRECVID 2006: Shot boundary detection task overview. In Proceedings of the TRECVID Workshop, November 2006.
• 10: P. Viola and M. Jones. Fast and robust classification using asymmetric adaboost and a detector cascade. In T. G. Dietterich, S. Becker, and Z. Ghahramani, editors, Advances in Neural Information Processing Systems 14, Cambridge, MA, 2002. MIT Press.
• 11: Michail Vlachos , Dimitrios Gunopoulos , George Kollios, Discovering Similar Multidimensional Trajectories, Proceedings of the 18th International Conference on Data Engineering, p.673, February 26-March 01, 2002

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.5 INFORMATION INTERFACES AND PRESENTATION (I.7)
      H.5.1 Multimedia Information Systems

Additional Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.1 Content Analysis and Indexing
          Subjects: Abstracting methods
      H.3.3 Information Search and Retrieval

General Terms:
Algorithms, Experimentation

Keywords:
content abstraction, hmm, rushes, summarization, trecvid, video skim