open-mmlab/OpenPCDet

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

5718 stars · 1473 forks observed · 2026-09-24

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

OpenPCDet is a PyTorch-based open-source toolbox for LiDAR-based 3D object detection. It provides official implementations of models like PointRCNN, PV-RCNN, Voxel R-CNN, MPPNet, and BEVFusion, with support for datasets such as Waymo, NuScenes, and Argoverse2.

Use cases

• train a 3D object detection model on LiDAR point clouds
• run 3D detection on Waymo or NuScenes datasets
• reproduce PV-RCNN or Voxel R-CNN research results
• fuse camera and LiDAR data with BEVFusion
• benchmark 3D detectors for autonomous driving
• deploy real-time LiDAR detection with TensorRT

When to choose

• you need state-of-the-art LiDAR 3D detection models in PyTorch
• you are doing autonomous driving perception research
• you want official implementations of PointRCNN, PV-RCNN, or MPPNet
• you need multi-modal (LiDAR + camera) detection support

When to avoid

• you need 2D image-only object detection
• you work outside Python/PyTorch ecosystems
• you need a production inference engine rather than a research toolbox
• you lack a GPU, since training requires significant compute

Facets

library · maturity active

machine-learning deep-learning computer-vision image-processing autonomous-vehicles computer-vision deep-learning machine-learning python lidar point-cloud 3d-object-detection pytorch pv-rcnn waymo nuscenes autonomous-driving linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="open-mmlab/OpenPCDet")