Scalable Surface Reconstruction with Delaunay-Graph Neural Networks
| dc.contributor.author | Sulzer, Raphael | en_US |
| dc.contributor.author | Landrieu, Loic | en_US |
| dc.contributor.author | Marlet, Renaud | en_US |
| dc.contributor.author | Vallet, Bruno | en_US |
| dc.contributor.editor | Digne, Julie and Crane, Keenan | en_US |
| dc.date.accessioned | 2021-07-10T07:46:24Z | |
| dc.date.available | 2021-07-10T07:46:24Z | |
| dc.date.issued | 2021 | |
| dc.description.abstract | We introduce a novel learning-based, visibility-aware, surface reconstruction method for large-scale, defect-laden point clouds. Our approach can cope with the scale and variety of point cloud defects encountered in real-life Multi-View Stereo (MVS) acquisitions. Our method relies on a 3D Delaunay tetrahedralization whose cells are classified as inside or outside the surface by a graph neural network and an energy model solvable with a graph cut. Our model, making use of both local geometric attributes and line-of-sight visibility information, is able to learn a visibility model from a small amount of synthetic training data and generalizes to real-life acquisitions. Combining the efficiency of deep learning methods and the scalability of energybased models, our approach outperforms both learning and non learning-based reconstruction algorithms on two publicly available reconstruction benchmarks. | en_US |
| dc.description.number | 5 | |
| dc.description.sectionheaders | Surface Reconstruction | |
| dc.description.seriesinformation | Computer Graphics Forum | |
| dc.description.volume | 40 | |
| dc.identifier.doi | 10.1111/cgf.14364 | |
| dc.identifier.issn | 1467-8659 | |
| dc.identifier.pages | 157-167 | |
| dc.identifier.uri | https://doi.org/10.1111/cgf.14364 | |
| dc.identifier.uri | https://diglib.eg.org:443/handle/10.1111/cgf14364 | |
| dc.publisher | The Eurographics Association and John Wiley & Sons Ltd. | en_US |
| dc.subject | Computing methodologies | |
| dc.subject | Reconstruction | |
| dc.subject | Neural networks | |
| dc.subject | Shape inference | |
| dc.title | Scalable Surface Reconstruction with Delaunay-Graph Neural Networks | en_US |