Realtime Anomaly Detection Using Trajectory-Level Crowd Behavior Learning
Abstract
We present an algorithm for realtime anomaly detection in low to medium density crowd videos using trajectorylevel behavior learning. Our formulation combines online tracking algorithms from computer vision, non-linear pedestrian motion models from crowd simulation, and Bayesian learning techniques to automatically compute the trajectory-level pedestrian behaviors for each agent in the video. These learned behaviors are used to segment the trajectories and motions of different pedestrians or agents and detect anomalies. We demonstrate the interactive performance on the PETS 2016 ARENA dataset as well as indoor and outdoor crowd video benchmarks consisting of tens of human agents.
Cite
Text
Bera et al. "Realtime Anomaly Detection Using Trajectory-Level Crowd Behavior Learning." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2016. doi:10.1109/CVPRW.2016.163Markdown
[Bera et al. "Realtime Anomaly Detection Using Trajectory-Level Crowd Behavior Learning." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2016.](https://mlanthology.org/cvprw/2016/bera2016cvprw-realtime/) doi:10.1109/CVPRW.2016.163BibTeX
@inproceedings{bera2016cvprw-realtime,
title = {{Realtime Anomaly Detection Using Trajectory-Level Crowd Behavior Learning}},
author = {Bera, Aniket and Kim, Sujeong and Manocha, Dinesh},
booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
year = {2016},
pages = {1289-1296},
doi = {10.1109/CVPRW.2016.163},
url = {https://mlanthology.org/cvprw/2016/bera2016cvprw-realtime/}
}