# sshaoshuai/PointRCNN

PointRCNN: 3D Object Proposal Generation and Detection from Point Cloud, CVPR 2019.

Repository: https://github.com/sshaoshuai/PointRCNN
Canonical: https://ross.abutalabs.com/products/pointrcnn
Language: Python
License: MIT
License Family: permissive
Last push: 2022-09-01T10:13:35+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2711, "days_push": 1462, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1817, forks 422 (observed 2026-08-28T04:05:40.126772+00:00)

## What it is
PyTorch implementation of PointRCNN, a two-stage 3D object detector that generates 3D box proposals directly from raw LiDAR point clouds. It is the official research code release for the CVPR 2019 paper, evaluated on the KITTI dataset.

## Use cases
- detect 3D objects from LiDAR point clouds
- train a 3D object detection model on KITTI
- reproduce PointRCNN CVPR 2019 results
- run 3D box proposal generation from raw point cloud
- benchmark two-stage point cloud detectors
- research bin-based 3D box regression losses

## When to choose
- you need the original PointRCNN implementation for research or paper reproduction
- you work with KITTI-format LiDAR data and PyTorch
- you want a reference two-stage point-cloud-only 3D detector

## When to avoid
- you need multi-class joint training or a maintained toolbox - use OpenPCDet instead
- you need Windows or production-ready deployment support
- you want actively updated code - the repo is a frozen research release

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, computer-vision
- domain: computer-vision, autonomous-vehicles, deep-learning
- platform: python
- tags: 3d-object-detection, point-cloud, lidar, pytorch, kitti, cvpr-2019, research-code, linux, gpu

## Member repositories
- sshaoshuai/PointRCNN (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:40.126772+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:20:34.277569+00:00, confidence not recorded.
  - readme: https://github.com/sshaoshuai/PointRCNN (fetched 2026-08-28T04:05:40.126772+00:00, sha 642893735be8)
- Data as of 2026-08-30T08:39:29.467469+00:00.