What question did this study set out to answer?

This research aims to enhance the efficiency of drone surveillance in detecting crowd anomalies using active perception techniques.

Visual feedback-driven active perception by drone swarms for proactive crowd anomaly capture

View Full Paper

Ask AI

Bookmark

Authors

Randomized trial demonstrates enhanced crowd anomaly capture in large environments, indicating improved surveillance efficacy.

This research aims to enhance the efficiency of drone surveillance in detecting crowd anomalies using active perception techniques.
Developed an active perception framework for drone swarms that utilizes real-time visual-semantic feedback.
Integrated a spatiotemporal graph attention network for modeling crowd interactions and inferring anomaly probabilities.
Evaluated the system on the VisDrone dataset to assess anomaly capture rates and operational efficiency.
Achieved an anomaly capture rate of 89.7%.
Demonstrated an average response delay of 1.9 seconds.
Achieved an operational efficiency of 1.86 events per kilometer flown with low observation redundancy of 22.1%.

Han et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark