Online Pedestrian Group Walking Event Detection Using Spectral Analysis of Motion Similarity Graph
This addresses the need for real-time group detection in surveillance, but it is incremental as it applies existing spectral clustering to a specific challenge dataset.
The paper tackled the problem of online detection of pedestrian groups in video by identifying objects with similar motion patterns using spectral clustering on a motion similarity graph, achieving evaluation on the PETS2015 dataset.
A method for online identification of group of moving objects in the video is proposed in this paper. This method at each frame identifies group of tracked objects with similar local instantaneous motion pattern using spectral clustering on motion similarity graph. Then, the output of the algorithm is used to detect the event of more than two object moving together as required by PETS2015 challenge. The performance of the algorithm is evaluated on the PETS2015 dataset.