# magicleap/SuperPointPretrainedNetwork

PyTorch pre-trained model for real-time interest point detection, description, and sparse tracking (https://arxiv.org/abs/1712.07629)

Repository: https://github.com/magicleap/SuperPointPretrainedNetwork
Canonical: https://ross.abutalabs.com/products/superpointpretrainednetwork
Language: Python
License: NOASSERTION
License Family: other
Last push: 2022-07-24T18:56:31+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3001, "days_push": 1501, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2185, forks 423 (observed 2026-08-28T04:06:23.653773+00:00)

## What it is
A PyTorch pre-trained implementation of the SuperPoint fully convolutional neural network for real-time interest point detection and descriptor computation, with a demo script for sparse optical flow tracking across image sequences. It includes a weights file and Python deployment script supporting image directories, video files, and webcam input.

## Use cases
- detect interest points in images with a pretrained neural network
- compute descriptors for image-to-image feature matching
- track sparse points across video frames for optical flow
- build visual SLAM or VO front-ends
- run feature detection on webcam or video files in real time

## When to choose
- you need a self-supervised learned alternative to hand-crafted feature detectors like SIFT or ORB
- you want a simple PyTorch script to evaluate SuperPoint on your own image or video data
- you are prototyping sparse point tracking for SLAM, VO, or image matching research

## When to avoid
- you need a maintained production library with API stability and active support
- you require dense optical flow rather than sparse point tracking
- you need a non-PyTorch framework or mobile/embedded deployment out of the box

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, computer-vision, image-processing
- domain: computer-vision, machine-learning, deep-learning
- platform: python
- tags: superpoint, feature-detection, interest-points, optical-flow, slam, pytorch, pretrained-model, feature-matching, linux, macos, gpu

## Member repositories
- magicleap/SuperPointPretrainedNetwork (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:23.653773+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 2026-08-30T02:48:11.849407+00:00, confidence not recorded.
  - readme: https://github.com/magicleap/SuperPointPretrainedNetwork (fetched 2026-08-28T04:06:23.653773+00:00, sha 35c367ca61c4)
- Data as of 2026-08-30T08:39:29.467469+00:00.