UMAD: Unsupervised Mask-Level Anomaly Detection for Autonomous Driving
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866929522136317952 |
|---|---|
| author | Bogdoll, Daniel Ollick, Noël Joseph, Tim Pavlitska, Svetlana Zöllner, J. Marius |
| author_facet | Bogdoll, Daniel Ollick, Noël Joseph, Tim Pavlitska, Svetlana Zöllner, J. Marius |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_06370 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | UMAD: Unsupervised Mask-Level Anomaly Detection for Autonomous Driving Bogdoll, Daniel Ollick, Noël Joseph, Tim Pavlitska, Svetlana Zöllner, J. Marius Computer Vision and Pattern Recognition Robotics 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. |
| title | UMAD: Unsupervised Mask-Level Anomaly Detection for Autonomous Driving |
| topic | Computer Vision and Pattern Recognition Robotics |
| url | https://arxiv.org/abs/2406.06370 |