3D Point Cloud Registration with Multi-Scale Architecture and Unsupervised transfer learning - Mines Paris
Article Dans Une Revue 2021 International Conference on 3D Vision (3DV) Année : 2021

3D Point Cloud Registration with Multi-Scale Architecture and Unsupervised transfer learning

Jean-Emmanuel Deschaud
Francois Goulette

Résumé

We present MS-SVConv, a fast multi-scale deep neural network that outputs features from point clouds for 3D registration between two scenes. We compute features using a 3D sparse voxel convolutional network on a point cloud at different scales and then fuse the features through fully-connected layers. With supervised learning, we show significant improvements compared to state-of-the-art methods on the competitive and well-known 3DMatch benchmark. We also achieve a better generalization through different source and target datasets, with very fast computation. Finally, we present a strategy to fine-tune MS-SVConv on unknown datasets in a self-supervised way, which leads to state-of-the-art results on ETH and TUM datasets.

Dates et versions

hal-03203390 , version 1 (20-04-2021)

Identifiants

Citer

Sofiane Horache, Jean-Emmanuel Deschaud, Francois Goulette. 3D Point Cloud Registration with Multi-Scale Architecture and Unsupervised transfer learning. 2021 International Conference on 3D Vision (3DV), 2021, ⟨10.1109/3DV53792.2021.00142⟩. ⟨hal-03203390⟩
150 Consultations
0 Téléchargements

Altmetric

Partager

More