Privacy preserving crowd monitoring: Counting people without people models or tracking

TLDR

The paper proposes a privacy‑preserving system that estimates crowd size in heterogeneous pedestrian flows without explicit segmentation or tracking. The system segments crowds into homogeneous motion components using a mixture of dynamic textures model, extracts holistic features from each segment, and maps these features to person counts with Gaussian Process regression, validated on a 2000‑frame pedestrian dataset. The system achieves accurate crowd size estimates when run on a full hour of video.

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.