Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2412.04120 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917967577481216 |
|---|---|
| author | Walker, Thomas Esposito, Salvatore Rebain, Daniel Vaxman, Amir Onken, Arno Li, Changjian Mac Aodha, Oisin |
| author_facet | Walker, Thomas Esposito, Salvatore Rebain, Daniel Vaxman, Amir Onken, Arno Li, Changjian Mac Aodha, Oisin |
| contents | Reconstructing complex structures from planar cross-sections is a challenging problem, with wide-reaching applications in medical imaging, manufacturing, and topography. Out-of-the-box point cloud reconstruction methods can often fail due to the data sparsity between slicing planes, while current bespoke methods struggle to reconstruct thin geometric structures and preserve topological continuity. This is important for medical applications where thin vessel structures are present in CT and MRI scans. This paper introduces CrossSDF, a novel approach for extracting a 3D signed distance field from 2D signed distances generated from planar contours. Our approach makes the training of neural SDFs contour-aware by using losses designed for the case where geometry is known within 2D slices. Our results demonstrate a significant improvement over existing methods, effectively reconstructing thin structures and producing accurate 3D models without the interpolation artifacts or over-smoothing of prior approaches. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_04120 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | CrossSDF: 3D Reconstruction of Thin Structures From Cross-Sections Walker, Thomas Esposito, Salvatore Rebain, Daniel Vaxman, Amir Onken, Arno Li, Changjian Mac Aodha, Oisin Computer Vision and Pattern Recognition Reconstructing complex structures from planar cross-sections is a challenging problem, with wide-reaching applications in medical imaging, manufacturing, and topography. Out-of-the-box point cloud reconstruction methods can often fail due to the data sparsity between slicing planes, while current bespoke methods struggle to reconstruct thin geometric structures and preserve topological continuity. This is important for medical applications where thin vessel structures are present in CT and MRI scans. This paper introduces CrossSDF, a novel approach for extracting a 3D signed distance field from 2D signed distances generated from planar contours. Our approach makes the training of neural SDFs contour-aware by using losses designed for the case where geometry is known within 2D slices. Our results demonstrate a significant improvement over existing methods, effectively reconstructing thin structures and producing accurate 3D models without the interpolation artifacts or over-smoothing of prior approaches. |
| title | CrossSDF: 3D Reconstruction of Thin Structures From Cross-Sections |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2412.04120 |