Efficient 3D Semantic Segmentation with Superpoint Transformer
Creators
- 1. LASTIG, IGN, ENSG, Univ Gustave Eiffel, France / CSAI, ENGIE Lab CRIGEN, France
- 2. INSA Centre Val-de-Loire Univ de Tours, LIFAT, France
- 3. LIGM, Ecole des Ponts, Univ Gustave Eiffel, CNRS, France / LASTIG, IGN, ENSG, Univ Gustave Eiffel, France
Contributors
Contact person:
- 1. LASTIG, IGN, ENSG, Univ Gustave Eiffel, France / CSAI, ENGIE Lab CRIGEN, France
Description
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.
Notes
Files
Files
(35.4 MB)
Name | Size | Download all |
---|---|---|
md5:c658329e6d1a60a746cef8c62ca94a1d
|
2.9 MB | Download |
md5:99a81f01d3ee2ecfc5f67f727a1d0e81
|
9.7 MB | Download |
md5:d7442d12b6986deaeb94465f7eb52d04
|
624.7 kB | Download |
md5:72c4742424712875c6502e6e259aa93b
|
1.2 MB | Download |
md5:b272075ab66fe37662a5c9ed4a3b32c0
|
633.5 kB | Download |
md5:16f20bdcf9559efaebb7305ce100c64d
|
633.5 kB | Download |
md5:a313b5c95ad661c07a9ce43f11c4828b
|
633.5 kB | Download |
md5:cf596ebd2a952acc00dcf2d58c953f8e
|
633.5 kB | Download |
md5:3b9bfa0b7631e907226b51a63dee57f4
|
633.5 kB | Download |
md5:ee6cf21143306f299acb056164789e04
|
633.5 kB | Download |
md5:02f820aa552b5395763c71898766bc07
|
2.9 MB | Download |
md5:92de023c861bc760fe92e13828baa4bc
|
2.9 MB | Download |
md5:d1cfa3fbc3b220826f4a8fbced85a530
|
2.9 MB | Download |
md5:1b371c197a1f753dad224d52e050a877
|
2.9 MB | Download |
md5:a8bedc73b3941574b62b17d36e6f302a
|
2.9 MB | Download |
md5:63f94ab88e9ab408848ad8fe79915c9f
|
2.9 MB | Download |