On-the-Fly Global Activity Prediction and Anomaly Detection

Abstract

We propose a unified framework using Latent Dirichlet Allocation (LDA) for discovering behaviour global correlations over a distributed camera network. We explore LDA for categorising object motion patterns as local behaviours in each camera view before correlating these local behaviours globally over different physical locations in multi-camera views. In particular, a Temporal Order Sensitive LDA (TOS-LDA) is formulated to discover behaviour global temporal correlations of different durations among all camera views simultaneously. In addition, a novel on-line global activity prediction method is proposed based on which global anomalies can be detected on the fly. We validate the effectiveness of our approach using public multi-camera CCTV footages.

Cite

Text

Li et al. "On-the-Fly Global Activity Prediction and Anomaly Detection." IEEE/CVF International Conference on Computer Vision Workshops, 2009. doi:10.1109/ICCVW.2009.5457455

Markdown

[Li et al. "On-the-Fly Global Activity Prediction and Anomaly Detection." IEEE/CVF International Conference on Computer Vision Workshops, 2009.](https://mlanthology.org/iccvw/2009/li2009iccvw-onthefly/) doi:10.1109/ICCVW.2009.5457455

BibTeX

@inproceedings{li2009iccvw-onthefly,
  title     = {{On-the-Fly Global Activity Prediction and Anomaly Detection}},
  author    = {Li, Jian and Gong, Shaogang and Xiang, Tao},
  booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
  year      = {2009},
  pages     = {1330-1337},
  doi       = {10.1109/ICCVW.2009.5457455},
  url       = {https://mlanthology.org/iccvw/2009/li2009iccvw-onthefly/}
}