In this video we present results for the task of unsupervised anomaly detection for aerial robotic surveillance. For environments for which anomaly data are sparse or absent altogether, this work proposes the merging of deep learned visual features and one-class support vector machines to efficiently detect anomaly on camera data and in real-time. Results are shown in relation to area surveillance using a camera-equipped aerial robot conducting a coverage path over an area in which few man-made structures are introduced and have the role of anomalies against their environment.
Training Data: Camera frames from similar environments but lacking any man-made structure or humans.
Test Data: Camera frames collected by the aerial robot over an area similar to that of the training data but in which also few man-made structures and humans have been introduced. Those man-made structures and humans should be detected as cases of anomaly in the data.
This preliminary work is presented as Late Breaking Result at IEEE ICRA 2018
Training Data: Camera frames from similar environments but lacking any man-made structure or humans.
Test Data: Camera frames collected by the aerial robot over an area similar to that of the training data but in which also few man-made structures and humans have been introduced. Those man-made structures and humans should be detected as cases of anomaly in the data.
This preliminary work is presented as Late Breaking Result at IEEE ICRA 2018