Efficient 3D Semantic Segmentation with Superpoint Transformer - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Efficient 3D Semantic Segmentation with Superpoint Transformer

Résumé

We introduce a novel superpoint-based transformer architecture for efficient semantic segmentation of large-scale 3D scenes. Our method incorporates a fast algorithm to partition point clouds into a hierarchical superpoint structure, which makes our preprocessing 7 times times faster than existing superpoint-based approaches. Additionally, we leverage a self-attention mechanism to capture the relationships between superpoints at multiple scales, leading to state-of-the-art performance on three challenging benchmark datasets: S3DIS (76.0% mIoU 6-fold validation), KITTI-360 (63.5% on Val), and DALES (79.6%). With only 212k parameters, our approach is up to 200 times more compact than other state-of-the-art models while maintaining similar performance. Furthermore, our model can be trained on a single GPU in 3 hours for a fold of the S3DIS dataset, which is 7x to 70x fewer GPU-hours than the best-performing methods. Our code and models are accessible at github.com/drprojects/superpoint_transformer.
Fichier principal
Vignette du fichier
2306.08045.pdf (8.43 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04140052 , version 1 (24-06-2023)

Licence

Paternité

Identifiants

Citer

Damien Robert, Hugo Raguet, Loic Landrieu. Efficient 3D Semantic Segmentation with Superpoint Transformer. International Conference on Computer Vision, CVR, IEEE, Oct 2023, Paris, France. pp.17195-17204. ⟨hal-04140052⟩
51 Consultations
65 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More