# facebookresearch/votenet

Deep Hough Voting for 3D Object Detection in Point Clouds

Repository: https://github.com/facebookresearch/votenet
Canonical: https://ross.abutalabs.com/products/votenet
Language: Python
License: MIT
License Family: permissive
Topics: 3d, object-detection, point-cloud, robotics, deep-learning, voting
Archived: true
Last push: 2022-01-30T06:12:55+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2574, "days_push": 1676, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1761, forks 388 (observed 2026-08-28T04:05:32.629696+00:00)

## What it is
VoteNet is the official PyTorch implementation of the ICCV 2019 paper 'Deep Hough Voting for 3D Object Detection in Point Clouds'. It provides an end-to-end 3D object detection network combining deep point set networks with Hough voting, plus data preparation, training, and evaluation scripts for SUN RGB-D and ScanNet datasets.

## Use cases
- detect 3D bounding boxes in point clouds
- train a 3D object detection model on ScanNet
- evaluate 3D detection on SUN RGB-D
- reproduce ICCV 2019 VoteNet paper results
- benchmark 3D detectors using only geometry without RGB images
- build robotics perception on indoor 3D scans

## When to choose
- you need pure-geometry 3D object detection from point clouds
- you want a compact, efficient reference model for ScanNet or SUN RGB-D
- you are doing research on point cloud detection and want a strong baseline

## When to avoid
- you need actively maintained code with modern PyTorch/CUDA versions
- you need RGB-image-based or LiDAR outdoor detection (e.g., KITTI/nuScenes)
- you want a production-ready detection pipeline rather than research code

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, computer-vision, image-processing
- domain: computer-vision, robotics, deep-learning, machine-learning
- platform: python
- tags: 3d-object-detection, point-cloud, hough-voting, pytorch, research-code, scannet, sun-rgbd, linux, gpu

## Member repositories
- facebookresearch/votenet (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:32.629696+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 2026-08-30T03:27:11.243492+00:00, confidence not recorded.
  - readme: https://github.com/facebookresearch/votenet (fetched 2026-08-28T04:05:32.629696+00:00, sha 8c42a0a57723)
- Data as of 2026-08-30T08:39:29.467469+00:00.