Loading...

Statistical Video Indexing

Roozgard, Amin Mohammad | 2008

637 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 39334 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Rabiee, Hamid Reza
  7. Abstract:
  8. Nowadays, video search and retrieval is interesting for computer users and it has chief usages for multimedia systems. Video generation rate has increased and Internet as a communication framework is case of its transferring on the world. Because of these, importance of video files is more than past. Searching for finding content will be faster if video files would have indexed with a comprehensive system. The biggest step in this way is power of index generation that would be same or similar to human mind, for improvement of the clustering’s result or classification’s result. For generating suitable indexes, it is necessary to extracting effective features from videos and synthesizing these features to make meaningful objects matched or semi-matched to human thinking. Statistical video modeling is one of solutions in this area. The various statistical methods have been using, for example, momentum based methods are very favorite between researchers. In this thesis, synthesis of primary features with Gaussian Mixture Model was used to reaching of content. For this purpose, video was analyzed based on spatio and spatio-tempo properties. First, we work to synthesizing video which is only based on local properties of each frame and used Gaussian Mixture Model(GMM) and Kullback–Leibler divergence (KLD) for indexing. Furthermore, we work to synthesizing video which based on local properties in time durations. We declared a new video entity that is very impacted and compressed. We represented video with this method and used GMM and KLD for modeling. The Results of our implementation confirmed the usability of our modeling
  9. Keywords:
  10. Indexing ; Gaussian Mixture Modeling ; Content Retrieval ; Video Files

 Digital Object List

  • محتواي پايان نامه
  •   view

 Bookmark

No TOC