# V2AI/Det3D

World's first general purpose 3D object detection codebse.

Repository: https://github.com/V2AI/Det3D
Canonical: https://ross.abutalabs.com/products/det3d
Homepage: https://arxiv.org/abs/1908.09492
Language: Python
License: Apache-2.0
License Family: permissive
Topics: nuscenes, 3d-object-detection, point-cloud, object-detection, kitti, pytorch, autonomous-driving, deep-learning
Last push: 2023-12-19T10:21:31+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": 2571, "days_push": 988, "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 1560, forks 291 (observed 2026-08-28T04:05:03.746899+00:00)

## What it is
Det3D is a PyTorch-based toolbox for 3D object detection from point clouds, offering implementations of models like PointPillars, SECOND, and VoxelNet. It supports major autonomous driving benchmarks including KITTI, nuScenes, and Lyft with pretrained model checkpoints.

## Use cases
- train 3d object detection models on nuscenes
- run pointpillars on kitti point clouds
- detect cars in lidar point clouds
- reproduce cbgs nuScenes benchmark results
- fine-tune a voxel-based 3d detector on my own dataset
- compare point-based and voxel-based 3d detection models
- distributed training of 3d detection networks

## When to choose
- you need off-the-shelf implementations of SECOND, PointPillars, or VoxelNet
- you work with KITTI, nuScenes, or Lyft point cloud datasets
- you want pretrained checkpoints and benchmark baselines for 3D detection research

## When to avoid
- you need recent models or active maintenance - the project has seen limited updates
- you need Waymo dataset support, which is still a TODO
- you want 2D image detection or camera-only perception

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, computer-vision, image-processing
- domain: autonomous-vehicles, computer-vision, deep-learning, machine-learning
- platform: python
- tags: 3d-object-detection, point-cloud, lidar, nuscenes, kitti, pytorch, model-zoo, autonomous-driving, linux, gpu

## Member repositories
- V2AI/Det3D (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:03.746899+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 2026-08-30T04:29:46.294823+00:00, confidence not recorded.
  - readme: https://github.com/V2AI/Det3D (fetched 2026-08-28T04:05:03.746899+00:00, sha db1c9cc5f83b)
  - homepage: https://arxiv.org/abs/1908.09492 (fetched 2026-08-29T11:29:33.514534+00:00, sha e54e3ebc7f0a)
  - site_page: https://info.arxiv.org/about/donate.html (fetched 2026-08-29T11:29:33.523724+00:00, sha cca9c3a11c56)
  - site_page: https://info.arxiv.org/about/ourmembers.html (fetched 2026-08-29T11:29:33.526916+00:00, sha 47cbc55ff1de)
  - site_page: https://info.arxiv.org/about (fetched 2026-08-29T11:29:33.528807+00:00, sha a1f16f915a9a)
  - site_page: https://info.arxiv.org/labs/index.html (fetched 2026-08-29T11:29:33.525351+00:00, sha b14a8d05a0ec)
- Data as of 2026-08-30T08:39:29.467469+00:00.