R.B.: Modelling crowd scenes for event detection

This paper is a postprint of a paper submitted to and accepted for publication in ICPR 2006 and is subject to IEEE copyright. This work presents an automatic technique for detection of abnormal events in crowds. Crowd behaviour is difficult to predict and might not be easily semantically translated. Moreover it is difficulty to track individuals in the crowd using state of the art tracking algorithms. Therefore we characterise crowd behaviour by observing the crowd opti-cal flow and use unsupervised feature extraction to encode normal crowd behaviour. The unsupervised feature extrac-tion applies spectral clustering to find the optimal number of models to represent normal motion patterns. The mo-tion 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 demonstrate the effectiveness of the approach for detecting crowd emergency scenarios.