auduno/clmtrackr

Health v2 · maintenance only

Adoption not part of the score

6500 stars · 1134 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

clmtrackr is a JavaScript library for fitting facial models to faces in videos or images using Constrained Local Models with regularized landmark mean-shift. It tracks faces in the browser and outputs coordinate positions of facial feature points, supporting applications like face substitution, masking, and emotion detection.

Use cases

• track facial landmarks in browser video
• detect face feature points in images with javascript
• build face substitution or masking effects in the browser
• realtime face deformation in web apps
• emotion detection from facial expressions
• overlay caricatures or masks on a webcam feed

When to choose

• you need client-side facial landmark tracking without a server
• you want a lightweight, MIT-licensed browser library for face tracking
• you need coordinate positions of facial features for creative or analytical effects

When to avoid

• you need state-of-the-art accuracy or modern deep-learning-based face tracking
• you need server-side or Node.js face tracking
• you need active development and recent updates - the last release was in 2020

Facets

library · maturity maintenance

computer-vision image-processing machine-learning computer-vision browser-extensions web-development browser face-tracking facial-landmarks constrained-local-models webgl emotion-detection face-masking face-recognition javascript web

Member repositories

For agents

markdown · JSON · MCP: product_card(name="auduno/clmtrackr")