google-ai-edge/mediapipe
Cross-platform, customizable ML solutions for live and streaming media.
MediaPipe provides ready-to-use solutions and tools for adding AI features to applications across mobile, web, and desktop. It takes live or streaming media (video, audio, text) and provides analyses like object detection, pose tracking, or text classification. Developers and product managers can use this to quickly build innovative machine learning experiences for their end-users.
34,104 stars. Actively maintained with 45 commits in the last 30 days.
Use this if you are a developer or product manager looking to quickly integrate on-device machine learning capabilities, such as real-time object or gesture recognition, into your applications.
Not ideal if you need extremely low-level control over every aspect of your machine learning models and don't wish to leverage pre-built solutions or a structured framework.
Stars
34,104
Forks
5,842
Language
C++
License
Apache-2.0
Category
Last pushed
Mar 12, 2026
Commits (30d)
45
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/google-ai-edge/mediapipe"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Recent Releases
Related frameworks
houmo-ai/mediapipe_plus
The purpose of this project is to apply mediapipe to more AI chips.
vidursatija/BlazePalm
PyTorch & CoreML implementation of MediaPipe Hands
pjbelo/mediapipe-js-demos
Google MediaPipe Javascript Demos (including live demos)
beenotung/mediapipe-nodejs
A Node.js library for running MediaPipe models that are typically browser-only. This package...
gokulnpc/Face-Mesh-Detection-Mediapipe
This project utilizes MediaPipe and OpenCV to perform real-time face mesh detection using a...