Privacy Preserving Crowd Monitoring: Counting People Without People Models or Tracking
Abstract
We present a privacy-preserving system for estimating the size of inhomogeneous crowds, composed of pedestrians that travel in different directions, without using explicit object segmentation or tracking. First, the crowd is segmented into components of homogeneous motion, using the mixture of dynamic textures motion model. Second, a set of simple holistic features is extracted from each segmented region, and the correspondence between features and the number of people per segment is learned with Gaussian Process regression. We validate both the crowd segmentation algorithm, and the crowd counting system, on a large pedestrian dataset (2000 frames of video, containing 49,885 total pedestrian instances). Finally, we present results of the system running on a full hour of video.
Cite
Text
Chan et al. "Privacy Preserving Crowd Monitoring: Counting People Without People Models or Tracking." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2008. doi:10.1109/CVPR.2008.4587569Markdown
[Chan et al. "Privacy Preserving Crowd Monitoring: Counting People Without People Models or Tracking." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2008.](https://mlanthology.org/cvpr/2008/chan2008cvpr-privacy/) doi:10.1109/CVPR.2008.4587569BibTeX
@inproceedings{chan2008cvpr-privacy,
title = {{Privacy Preserving Crowd Monitoring: Counting People Without People Models or Tracking}},
author = {Chan, Antoni B. and Liang, Zhang-Sheng John and Vasconcelos, Nuno},
booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
year = {2008},
doi = {10.1109/CVPR.2008.4587569},
url = {https://mlanthology.org/cvpr/2008/chan2008cvpr-privacy/}
}