Multi-View Consistent Wound Segmentation With Neural Fields
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909998864400384 |
|---|---|
| author | Chierchia, Remi Lebrat, Léo Ahmedt-Aristizabal, David Arzhaeva, Yulia Salvado, Olivier Fookes, Clinton Cruz, Rodrigo Santa |
| author_facet | Chierchia, Remi Lebrat, Léo Ahmedt-Aristizabal, David Arzhaeva, Yulia Salvado, Olivier Fookes, Clinton Cruz, Rodrigo Santa |
| contents | Wound care is often challenged by the economic and logistical burdens that consistently afflict patients and hospitals worldwide. In recent decades, healthcare professionals have sought support from computer vision and machine learning algorithms. In particular, wound segmentation has gained interest due to its ability to provide professionals with fast, automatic tissue assessment from standard RGB images. Some approaches have extended segmentation to 3D, enabling more complete and precise healing progress tracking. However, inferring multi-view consistent 3D structures from 2D images remains a challenge. In this paper, we evaluate WoundNeRF, a NeRF SDF-based method for estimating robust wound segmentations from automatically generated annotations. We demonstrate the potential of this paradigm in recovering accurate segmentations by comparing it against state-of-the-art Vision Transformer networks and conventional rasterisation-based algorithms. The code will be released to facilitate further development in this promising paradigm. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_16487 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Multi-View Consistent Wound Segmentation With Neural Fields Chierchia, Remi Lebrat, Léo Ahmedt-Aristizabal, David Arzhaeva, Yulia Salvado, Olivier Fookes, Clinton Cruz, Rodrigo Santa Computer Vision and Pattern Recognition Wound care is often challenged by the economic and logistical burdens that consistently afflict patients and hospitals worldwide. In recent decades, healthcare professionals have sought support from computer vision and machine learning algorithms. In particular, wound segmentation has gained interest due to its ability to provide professionals with fast, automatic tissue assessment from standard RGB images. Some approaches have extended segmentation to 3D, enabling more complete and precise healing progress tracking. However, inferring multi-view consistent 3D structures from 2D images remains a challenge. In this paper, we evaluate WoundNeRF, a NeRF SDF-based method for estimating robust wound segmentations from automatically generated annotations. We demonstrate the potential of this paradigm in recovering accurate segmentations by comparing it against state-of-the-art Vision Transformer networks and conventional rasterisation-based algorithms. The code will be released to facilitate further development in this promising paradigm. |
| title | Multi-View Consistent Wound Segmentation With Neural Fields |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.16487 |