A Novel Event-Oriented Segment-of-Interest Discovery Method for Surveillance Video

Peng Cui, Lifeng Sun, Zhi Wang, Shi-Qiang Yang

Abstract

During recent years, the quick development of computer techniques has witnessed the ever-increasing surveillance video data, which essentially pose great challenge on the data storage, management, analysis and even retrieval. Considering that most of the high volume of data is with no interest, we mainly investigate the problem of effectively and efficiently discovering segments-of-interest (SoI) in this paper. To do so, we propose a novel event-oriented SoI discovery method in two steps: first, we represent an event by modeling pixels' change in temporal-spatial space, aiming to unify both the inter-frames and frames-background changes; second, with the benefit of unsupervised learning, the prototype-event models could be learned from these detected events and in turn exploited to measure the interest factor of each prototypeevent. The experiment results demonstrate that the proposed method precisely discriminate different events and effectively discover SoIs.
BibTeX
@inproceedings{cui2007novel,
  author = {Peng Cui and Lifeng Sun and Zhi Wang and Shi-Qiang Yang},
  title = {{A Novel Event-Oriented Segment-of-Interest Discovery Method for Surveillance Video}},
  booktitle = {ICME},
  year = {2007},
  doi = {10.1109/icme.2007.4284777},
}

Loading related content...

Figure