TR2004-061

Effective and Efficient Sports Highlights Extraction Using the Minimum Description Length Criterion in Selecting GMM Structures

    •  Xiong, Z., Radhakrishnan, R., Divakaran, A., Huang, T.S., "Effective and Efficient Sports Highlights Extraction Using the Minimum Description Length Criterion in Selecting GMM Structures", IEEE International Conference on Multimedia and Expo (ICME), June 2004.
      BibTeX TR2004-061 PDF
      • @inproceedings{Xiong2004jun,
      • author = {Xiong, Z. and Radhakrishnan, R. and Divakaran, A. and Huang, T.S.},
      • title = {Effective and Efficient Sports Highlights Extraction Using the Minimum Description Length Criterion in Selecting GMM Structures},
      • booktitle = {IEEE International Conference on Multimedia and Expo (ICME)},
      • year = 2004,
      • month = jun,
      • url = {https://www.merl.com/publications/TR2004-061}
      • }
  • Research Areas:

    Artificial Intelligence, Speech & Audio

Abstract:

In fitting the training data with Guassian Mixture Models (GMMs) or appropriate structures using the MDL criterion, we are able to improve audio classification accuracy with a large margin. With the MDL-GMMs, we are also able to greatly improve the accuracy in extracting sports highlights. Since we have focused on audio domain processing, it enables us to extract highlights very fast. In this paper, we have demonstrated the importance of a better understanding of model structures in such a pattern recognition task.

 

  • Related News & Events

    •  NEWS    ICME 2004: 5 publications by Anthony Vetro, Ajay Divakaran and Huifang Sun
      Date: June 27, 2004
      Where: IEEE International Conference on Multimedia and Expo (ICME)
      MERL Contacts: Anthony Vetro; Huifang Sun
      Brief
      • The papers "Adaptive Fast Playback-Based Video Skimming Using a Compressed-Domain Visual Complexity Measure" by Peker, K.A. and Divakaran, A., "Effective and Efficient Sports Highlights Extraction Using the Minimum Description Length Criterion in Selecting GMM Structures" by Xiong, Z., Radhakrishnan, R., Divakaran, A. and Huang, T.S., "Time Series Analysis and Segmentation Using Eigenvectors for Mining Semantic Audio Label Sequences" by Radhakrishnan, R., Xiong, Z., Divakaran, A. and Memon, N., "Towards Maximizing the End-User Experience" by Divakaran, A., Vetro, A. and Kan, T. and "Coding Artifact Reduction Using Edge Map Guided Adaptive and Fuzzy Filter" by Kong, H.-S., Nie, Y., Vetro, A., Sun, H. and Barner, K.E. were presented at the IEEE International Conference on Multimedia and Expo (ICME).
    •