Efficient 3D Semantic Segmentation with Superpoint Transformer

Damien Robert, Hugo Raguet, Loic Landrieu · ICCV 2023 · 2023

Status: To read
Topics: backbones
Feeds: First paper · height sampling

What they did

Superpoint Transformer partitions a point cloud into a hierarchical superpoint structure quickly and applies self-attention between superpoints at several scales, reaching state-of-the-art accuracy on S3DIS, KITTI-360 and DALES with only 212k parameters.

My note

No note yet.

Code

Repository: drprojects/superpoint_transformer
Framework: PyTorch
Licence: MIT
My status: Not tried

BibTeX

@inproceedings{robert2023-spt,
  title = {{Efficient 3D Semantic Segmentation with Superpoint Transformer}},
  author = {Damien Robert and Hugo Raguet and Loic Landrieu},
  booktitle = {ICCV 2023},
  year = {2023},
  eprint = {2306.08045},
  archivePrefix = {arXiv}
}