COARSE: Collaborative Pseudo-Labeling with Coarse Real Labels for Off-Road Semantic Segmentation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866908787121586176 |
|---|---|
| author | Noca, Aurelio Lei, Xianmei Becktor, Jonathan Edlund, Jeffrey Sabel, Anna Spieler, Patrick Padgett, Curtis Alahi, Alexandre Atha, Deegan |
| author_facet | Noca, Aurelio Lei, Xianmei Becktor, Jonathan Edlund, Jeffrey Sabel, Anna Spieler, Patrick Padgett, Curtis Alahi, Alexandre Atha, Deegan |
| contents | Autonomous off-road navigation faces challenges due to diverse, unstructured environments, requiring robust perception with both geometric and semantic understanding. However, scarce densely labeled semantic data limits generalization across domains. Simulated data helps, but introduces domain adaptation issues. We propose COARSE, a semi-supervised domain adaptation framework for off-road semantic segmentation, leveraging sparse, coarse in-domain labels and densely labeled out-of-domain data. Using pretrained vision transformers, we bridge domain gaps with complementary pixel-level and patch-level decoders, enhanced by a collaborative pseudo-labeling strategy on unlabeled data. Evaluations on RUGD and Rellis-3D datasets show significant improvements of 9.7\% and 8.4\% respectively, versus only using coarse data. Tests on real-world off-road vehicle data in a multi-biome setting further demonstrate COARSE's applicability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_03947 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | COARSE: Collaborative Pseudo-Labeling with Coarse Real Labels for Off-Road Semantic Segmentation Noca, Aurelio Lei, Xianmei Becktor, Jonathan Edlund, Jeffrey Sabel, Anna Spieler, Patrick Padgett, Curtis Alahi, Alexandre Atha, Deegan Computer Vision and Pattern Recognition Artificial Intelligence Robotics Autonomous off-road navigation faces challenges due to diverse, unstructured environments, requiring robust perception with both geometric and semantic understanding. However, scarce densely labeled semantic data limits generalization across domains. Simulated data helps, but introduces domain adaptation issues. We propose COARSE, a semi-supervised domain adaptation framework for off-road semantic segmentation, leveraging sparse, coarse in-domain labels and densely labeled out-of-domain data. Using pretrained vision transformers, we bridge domain gaps with complementary pixel-level and patch-level decoders, enhanced by a collaborative pseudo-labeling strategy on unlabeled data. Evaluations on RUGD and Rellis-3D datasets show significant improvements of 9.7\% and 8.4\% respectively, versus only using coarse data. Tests on real-world off-road vehicle data in a multi-biome setting further demonstrate COARSE's applicability. |
| title | COARSE: Collaborative Pseudo-Labeling with Coarse Real Labels for Off-Road Semantic Segmentation |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Robotics |
| url | https://arxiv.org/abs/2503.03947 |