jeeliz/jeelizFaceFilter

Health v2 · maintenance only

Adoption not part of the score

2928 stars · 546 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A lightweight JavaScript/WebGL library for real-time face detection and tracking from a camera feed via WebRTC, designed for building augmented reality face filters. It outputs raw face position, scale, and rotation data and integrates with 3D engines like Three.js and Babylon.js.

Use cases

• add snapchat-style face filters to a web app
• track face position and rotation in real time from webcam
• overlay 3D glasses or masks on a user's face in the browser
• build a virtual try-on experience for eyewear
• detect multiple faces simultaneously in a video feed
• detect mouth opening for avatar animation
• swap faces in real time in the browser

When to choose

• you need a lightweight, framework-agnostic face tracking library that runs in the browser
• you want to build AR face filters with Three.js, Babylon.js, or Canvas2D
• you need multi-face tracking without heavy dependencies
• you want a permissively licensed (Apache-2.0) solution for web AR

When to avoid

• you need server-side or native mobile face tracking
• you require full body or hand tracking rather than faces
• you need a complete turnkey filter platform rather than a tracking library
• your target browsers lack WebGL support

Facets

library · maturity active

computer-vision machine-learning graphics image-processing computer-vision web-development artificial-intelligence browser cross-platform face-tracking face-detection face-filters augmented-reality webgl webrtc threejs snapchat-style-filters real-time lightweight javascript

Member repositories

For agents

markdown · JSON · MCP: product_card(name="jeeliz/jeelizFaceFilter")