Multi-modality Anomaly Segmentation on the Road
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910889517514752 |
|---|---|
| author | Gao, Heng He, Zhuolin Qiu, Shoumeng Xue, Xiangyang Pu, Jian |
| author_facet | Gao, Heng He, Zhuolin Qiu, Shoumeng Xue, Xiangyang Pu, Jian |
| contents | Semantic segmentation allows autonomous driving cars to understand the surroundings of the vehicle comprehensively. However, it is also crucial for the model to detect obstacles that may jeopardize the safety of autonomous driving systems. Based on our experiments, we find that current uni-modal anomaly segmentation frameworks tend to produce high anomaly scores for non-anomalous regions in images. Motivated by this empirical finding, we develop a multi-modal uncertainty-based anomaly segmentation framework, named MMRAS+, for autonomous driving systems. MMRAS+ effectively reduces the high anomaly outputs of non-anomalous classes by introducing text-modal using the CLIP text encoder. Indeed, MMRAS+ is the first multi-modal anomaly segmentation solution for autonomous driving. Moreover, we develop an ensemble module to further boost the anomaly segmentation performance. Experiments on RoadAnomaly, SMIYC, and Fishyscapes validation datasets demonstrate the superior performance of our method. The code is available in https://github.com/HengGao12/MMRAS_plus. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_17712 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Multi-modality Anomaly Segmentation on the Road Gao, Heng He, Zhuolin Qiu, Shoumeng Xue, Xiangyang Pu, Jian Computer Vision and Pattern Recognition Artificial Intelligence Semantic segmentation allows autonomous driving cars to understand the surroundings of the vehicle comprehensively. However, it is also crucial for the model to detect obstacles that may jeopardize the safety of autonomous driving systems. Based on our experiments, we find that current uni-modal anomaly segmentation frameworks tend to produce high anomaly scores for non-anomalous regions in images. Motivated by this empirical finding, we develop a multi-modal uncertainty-based anomaly segmentation framework, named MMRAS+, for autonomous driving systems. MMRAS+ effectively reduces the high anomaly outputs of non-anomalous classes by introducing text-modal using the CLIP text encoder. Indeed, MMRAS+ is the first multi-modal anomaly segmentation solution for autonomous driving. Moreover, we develop an ensemble module to further boost the anomaly segmentation performance. Experiments on RoadAnomaly, SMIYC, and Fishyscapes validation datasets demonstrate the superior performance of our method. The code is available in https://github.com/HengGao12/MMRAS_plus. |
| title | Multi-modality Anomaly Segmentation on the Road |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2503.17712 |