QingyongHu/RandLA-Net

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1560 stars · 336 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Official TensorFlow implementation of RandLA-Net, a neural architecture for efficient semantic segmentation of large-scale 3D point clouds, published at CVPR 2020 (Oral) and IEEE TPAMI 2021. It includes training and evaluation scripts for benchmarks like S3DIS, Semantic3D, and SemanticKITTI with pre-trained models available.

Use cases

• segment large-scale 3d point clouds semantically
• run semantic segmentation on semantickitti dataset
• evaluate point cloud segmentation on s3dis benchmark
• reproduce cvpr 2020 randla-net results
• get pre-trained model for 3d lidar scene labeling
• benchmark efficient point cloud segmentation architectures

When to choose

• you need efficient semantic segmentation of large-scale 3D point clouds
• you want to reproduce or build on the RandLA-Net paper results
• you work with benchmarks like S3DIS, Semantic3D, or SemanticKITTI
• you need a lightweight point cloud segmentation model without heavy sampling or preprocessing

When to avoid

• you need PyTorch or modern TensorFlow 2.x support
• you require a permissively licensed model for commercial use (CC BY-NC-SA 4.0)
• you need actively maintained code with recent dependency updates
• your project involves 2D image segmentation rather than 3D point clouds

Facets

library · maturity maintenance

machine-learning deep-learning computer-vision computer-vision machine-learning deep-learning python point-clouds semantic-segmentation tensorflow cvpr-2020 3d-vision research-code linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="QingyongHu/RandLA-Net")