Visual Information Analysis for Big-Data Using Multi-core Technologies
The exponential growth of video data produced by surveillance cameras, cell phones and movie post-production creates the need to process big-data using methods that are able to produce instantaneous result. Video summarization can be accomplished and represented in several manners. The achieved summaries might be a sequence of images or short videos. In our method, an input video is divided into segments. From each segment we calculate key frames using three different key frame definitions, to summarize the video data. The contribution of this paper is to describe how to incorporate techniques that extract on the fly results.
Keywordsvideo summarization big-data video analysis mutual information
Unable to display preview. Download preview PDF.
- 3.Cernekova, Z., Pitas, I., Nikou, C.: Information theory-based shot cut/fade detection and video summarization. IEEE Transactions on Circuits and Systems for Video Technology 16 (January 2006)Google Scholar
- 4.Opencv metrics for histograms, http://docs.opencv.org/modules/imgproc/doc/histograms.html?highlight=comparehist#comparehist
- 6.Cernekova, Z., Kotropoulos, C., Pitas, I.: Video shot segmentation using singular value decomposition. SPIE Journal of Electronic Imaging 16(4) (December 2007)Google Scholar
- 7.Chen, Y.K., Holliman, M., Debes, E., Zheltov, S., Knyazev, A., Bratanov, S., ... Santos, I.: Media Applications on Hyper-Threading Technology. Journal Intel Technology 6(1) (2002)Google Scholar
- 8.Pitas, I., Venetsanopoulos, A.: Nonlinear Digital Filters: Principles and Applications. Kluwer Academic (1990)Google Scholar