Hypergraph Partitioning for Video Event Detection

With the development of computer technology, video event detection is becoming more and more important part of Recognition Pattern. Video event detection is the process of describing, identifying, classifying and interpreting the Objects or phenomena in the video. Finally across the underlying semantic gap, get high-level semantic concept. This paper improves the hypergraph partitioning theory, for the classification of the video events. The graph transformation tool AGG is used to conduct experiments in order to evaluate the accuracy of hypergraph method, a good result is obtained.