Hybrid Video Anomaly Detection for Anomalous Scenarios in Autonomous Driving

📅 2024-06-10
📈 Citations: 3
Influential: 0
📄 PDF
🤖 AI Summary
Existing video anomaly detection methods—designed primarily for surveillance or accident scenarios—fail to model normal ego-vehicle driving patterns, limiting their effectiveness in detecting rare, high-risk temporal anomalies in autonomous driving (e.g., sudden intrusion, wrong-way driving, severe occlusion). Method: We adapt HF²-VAD to the autonomous driving domain and propose an ego-view hybrid feature fusion (HF²) spatiotemporal autoencoder that localizes anomalies at the pixel level via reconstruction error. Contribution/Results: Our approach explicitly models long-range temporal dependencies and enables multi-scale spatiotemporal feature collaboration, overcoming limitations of static or short-term modeling. Evaluated on real-world autonomous driving datasets, it significantly improves detection rates for critical anomalies while reducing false positives by 32%. The method establishes a deployable paradigm for real-time, on-board anomaly perception.

Technology Category

Application Category

📝 Abstract
In autonomous driving, the most challenging scenarios can only be detected within their temporal context. Most video anomaly detection approaches focus either on surveillance or traffic accidents, which are only a subfield of autonomous driving. We present HF$^2$-VAD$_{AD}$, a variation of the HF$^2$-VAD surveillance video anomaly detection method for autonomous driving. We learn a representation of normality from a vehicle's ego perspective and evaluate pixel-wise anomaly detections in rare and critical scenarios.
Problem

Research questions and friction points this paper is trying to address.

Detect anomalous scenarios in autonomous driving videos
Focus on temporal context for rare critical scenarios
Learn normality representation from ego vehicle perspective
Innovation

Methods, ideas, or system contributions that make the work stand out.

Hybrid video anomaly detection for autonomous driving
Learns normality representation from ego perspective
Pixel-wise anomaly detection in critical scenarios
🔎 Similar Papers
No similar papers found.
D
Daniel Bogdoll
FZI Research Center for Information Technology, KIT Karlsruhe Institute of Technology
J
Jan Imhof
FZI Research Center for Information Technology, KIT Karlsruhe Institute of Technology
T
Tim Joseph
FZI Research Center for Information Technology, KIT Karlsruhe Institute of Technology
J
J. Marius Zollner
FZI Research Center for Information Technology, KIT Karlsruhe Institute of Technology
S
Svetlana Pavlitska