Learning Multimodal Deep Representations for Crowd Anomaly Event Detection

Explore this paper's citation graph

Summary

In this study, a novel unsupervised deep learning framework is proposed to detect anomaly events in crowded scenes using low-level visual features, energy features, and motion map features simultaneously extracted based on spatiotemporal energy measurements.

Type
article
Published
2018-01-31
Cited by
25
References
38
Access
Open access

Keywords

Anomaly detection, Artificial intelligence, Computer science, Boltzmann machine, Representation (politics)

References

Cited by

Related papers