Masked Gamma-SSL: Learning Uncertainty Estimation via Masked Image Modeling
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866929258003169280 |
|---|---|
| author | Williams, David S. W. Gadd, Matthew Newman, Paul De Martini, Daniele |
| author_facet | Williams, David S. W. Gadd, Matthew Newman, Paul De Martini, Daniele |
| contents | This work proposes a semantic segmentation network that produces high-quality uncertainty estimates in a single forward pass. We exploit general representations from foundation models and unlabelled datasets through a Masked Image Modeling (MIM) approach, which is robust to augmentation hyper-parameters and simpler than previous techniques. For neural networks used in safety-critical applications, bias in the training data can lead to errors; therefore it is crucial to understand a network's limitations at run time and act accordingly. To this end, we test our proposed method on a number of test domains including the SAX Segmentation benchmark, which includes labelled test data from dense urban, rural and off-road driving domains. The proposed method consistently outperforms uncertainty estimation and Out-of-Distribution (OoD) techniques on this difficult benchmark. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_17622 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Masked Gamma-SSL: Learning Uncertainty Estimation via Masked Image Modeling Williams, David S. W. Gadd, Matthew Newman, Paul De Martini, Daniele Computer Vision and Pattern Recognition Robotics This work proposes a semantic segmentation network that produces high-quality uncertainty estimates in a single forward pass. We exploit general representations from foundation models and unlabelled datasets through a Masked Image Modeling (MIM) approach, which is robust to augmentation hyper-parameters and simpler than previous techniques. For neural networks used in safety-critical applications, bias in the training data can lead to errors; therefore it is crucial to understand a network's limitations at run time and act accordingly. To this end, we test our proposed method on a number of test domains including the SAX Segmentation benchmark, which includes labelled test data from dense urban, rural and off-road driving domains. The proposed method consistently outperforms uncertainty estimation and Out-of-Distribution (OoD) techniques on this difficult benchmark. |
| title | Masked Gamma-SSL: Learning Uncertainty Estimation via Masked Image Modeling |
| topic | Computer Vision and Pattern Recognition Robotics |
| url | https://arxiv.org/abs/2402.17622 |