Counting people from above: Airborne video based crowd analysis
This provides a robust solution for security-critical crowd analysis in mass events, though it is incremental as it builds on existing object detection and regression techniques.
The paper tackled the problem of accurately estimating human count, density, and motion from airborne video for crowd monitoring in mass events, achieving a mean error of 4% to 9% in human counts.
Crowd monitoring and analysis in mass events are highly important technologies to support the security of attending persons. Proposed methods based on terrestrial or airborne image/video data often fail in achieving sufficiently accurate results to guarantee a robust service. We present a novel framework for estimating human count, density and motion from video data based on custom tailored object detection techniques, a regression based density estimate and a total variation based optical flow extraction. From the gathered features we present a detailed accuracy analysis versus ground truth measurements. In addition, all information is projected into world coordinates to enable a direct integration with existing geo-information systems. The resulting human counts demonstrate a mean error of 4% to 9% and thus represent a most efficient measure that can be robustly applied in security critical services.