UMAD: Unsupervised Mask-Level Anomaly Detection for Autonomous Driving
This addresses the scalability issue in anomaly detection for autonomous driving by eliminating the need for supervised data, though it appears incremental as it builds on existing generative and segmentation techniques.
The paper tackles the problem of detecting atypical traffic scenarios in autonomous driving without requiring labeled outlier data, presenting UMAD which outperforms state-of-the-art unsupervised anomaly detection methods.
Dealing with atypical traffic scenarios remains a challenging task in autonomous driving. However, most anomaly detection approaches cannot be trained on raw sensor data but require exposure to outlier data and powerful semantic segmentation models trained in a supervised fashion. This limits the representation of normality to labeled data, which does not scale well. In this work, we revisit unsupervised anomaly detection and present UMAD, leveraging generative world models and unsupervised image segmentation. Our method outperforms state-of-the-art unsupervised anomaly detection.