Performance analysis of event detection models in crowded scenes

Abstract

This paper evaluates an automatic technique for detection of abnormal events in crowds. Crowd behavior is difficult to predict and might not be easily semantically translated. Moreover it is difficult to track individuals in the crowd using state of the art tracking algorithms. Therefore we characterize crowd behavior by observing the crowd optical flow and use unsupervised feature extraction to encode normal crowd behavior. The unsupervised feature extraction applies spectral clustering to find the optimal number of models to represent normal motion patterns. The motion models are HMMs to cope with the variable number of motion samples that might be present in each observation window. The results on simulated crowds analyze the robustness of the approach for detecting crowd emergency scenarios observing the crowd at local and global levels. The results on normal real data show the effectiveness in modeling the more diverse behavior present in normal crowds. These results improve our previous work in Andrade, E.L. et al, (2005) in the detection of anomalies in pedestrian data.