# JEOresearch/EyeTracker

A lightweight and robust Python eye tracker

Repository: https://github.com/JEOresearch/EyeTracker
Canonical: https://ross.abutalabs.com/products/eyetracker
Language: Python
License: MIT
License Family: permissive
Last push: 2026-07-29T20:13:07+00:00

## Health v2 (maintenance only)
Score: 67/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 95, release rhythm 35, longevity 60
- inputs: {"age_days": 852, "days_push": 35, "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 1106, forks 169 (observed 2026-08-28T04:03:36.580349+00:00)

## What it is
A lightweight open-source Python library for 3D eye tracking that detects and fits the pupil ellipse in eye camera video or images. It is an optimized Python rewrite of the Yuta Itoh 3D eye tracker pupil fitter, with a faster lite variant included.

## Use cases
- track pupil position in eye camera video
- detect pupil ellipse in images with opencv
- build a DIY eye tracker for research
- integrate gaze tracking into a VR or AR headset prototype
- process eye recordings to extract pupil geometry
- experiment with 3D pupil fitting algorithms in Python

## When to choose
- you need a simple, hackable Python pupil tracker you can modify
- you have IR eye camera footage and want robust pupil detection without heavy dependencies
- you want a lightweight alternative to full eye-tracking SDKs

## When to avoid
- you need a complete gaze estimation or calibration pipeline out of the box
- you need real-time performance in a production product without tuning
- your environment requires numpy 2.x, which has a known incompatibility

## Facets
- artifact type: library
- maturity: active
- function: computer-vision, image-processing
- domain: computer-vision
- platform: python, cross-platform
- tags: eye-tracking, pupil-detection, opencv, 3d-eye-model, gaze-estimation, research

## Member repositories
- JEOresearch/EyeTracker (main) score 67

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:36.580349+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 2026-08-30T06:44:08.258650+00:00, confidence not recorded.
  - readme: https://github.com/JEOresearch/EyeTracker (fetched 2026-08-28T04:03:36.580349+00:00, sha 7f5586817271)
- Data as of 2026-08-30T08:39:29.467469+00:00.