SuperPoint Transformer
Performs semantic and panoptic segmentation of very large scans by grouping points before classifying them.
- Field
- Cross-domain
- Method
- Deep learning
- Data it takes
- point-cloud
- License
- MIT
Rather than classifying every point independently, this groups geometrically similar points into superpoints and reasons over those. The reduction makes scans tractable that would otherwise be too large, and it produces panoptic output: both what each region is and which object instance it belongs to.
Suited to survey-scale scans where per-point methods become impractical.
Catalog entry last checked 2026-08-19. All models →
Run SuperPoint Transformer on your data
The lab runs this on its own compute, in a container, with the inputs and parameters recorded alongside the result. Initial scoping conversations are free.
Contact the Lab