Tree-based Semantic Losses: Application to Sparsely-supervised Large Multi-class Hyperspectral Segmentation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911193652789248 |
|---|---|
| author | Wang, Junwen Maccormac, Oscar Rochford, William Kujawa, Aaron Shapey, Jonathan Vercauteren, Tom |
| author_facet | Wang, Junwen Maccormac, Oscar Rochford, William Kujawa, Aaron Shapey, Jonathan Vercauteren, Tom |
| contents | Hyperspectral imaging (HSI) shows great promise for surgical applications, offering detailed insights into biological tissue differences beyond what the naked eye can perceive. Refined labelling efforts are underway to train vision systems to distinguish large numbers of subtly varying classes. However, commonly used learning methods for biomedical segmentation tasks penalise all errors equivalently and thus fail to exploit any inter-class semantics in the label space. In this work, we introduce two tree-based semantic loss functions which take advantage of a hierarchical organisation of the labels. We further incorporate our losses in a recently proposed approach for training with sparse, background-free annotations. Extensive experiments demonstrate that our proposed method reaches state-of-the-art performance on a sparsely annotated HSI dataset comprising $107$ classes organised in a clinically-defined semantic tree structure. Furthermore, our method enables effective detection of out-of-distribution (OOD) pixels without compromising segmentation performance on in-distribution (ID) pixels. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_21150 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Tree-based Semantic Losses: Application to Sparsely-supervised Large Multi-class Hyperspectral Segmentation Wang, Junwen Maccormac, Oscar Rochford, William Kujawa, Aaron Shapey, Jonathan Vercauteren, Tom Computer Vision and Pattern Recognition Hyperspectral imaging (HSI) shows great promise for surgical applications, offering detailed insights into biological tissue differences beyond what the naked eye can perceive. Refined labelling efforts are underway to train vision systems to distinguish large numbers of subtly varying classes. However, commonly used learning methods for biomedical segmentation tasks penalise all errors equivalently and thus fail to exploit any inter-class semantics in the label space. In this work, we introduce two tree-based semantic loss functions which take advantage of a hierarchical organisation of the labels. We further incorporate our losses in a recently proposed approach for training with sparse, background-free annotations. Extensive experiments demonstrate that our proposed method reaches state-of-the-art performance on a sparsely annotated HSI dataset comprising $107$ classes organised in a clinically-defined semantic tree structure. Furthermore, our method enables effective detection of out-of-distribution (OOD) pixels without compromising segmentation performance on in-distribution (ID) pixels. |
| title | Tree-based Semantic Losses: Application to Sparsely-supervised Large Multi-class Hyperspectral Segmentation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2506.21150 |