mediapipe

Cross-platform, customizable ML solutions for live and streaming media.

What is mediapipe?

Cross-platform, customizable ML solutions for live and streaming media.

Overview of mediapipe

What it is

MediaPipe is a cross-platform, customizable machine learning solution for live and streaming media. It provides on-device ML capabilities for mobile (Android, iOS), web, desktop, edge devices, and IoT. The project includes MediaPipe Solutions (pre-built libraries and tools), MediaPipe Framework (low-level pipeline components), and MediaPipe Studio (a browser-based tool for visualization and benchmarking).

Key features

  • Cross-platform support: Android, iOS, web, desktop, edge devices, and IoT - Pre-trained, ready-to-run models for various tasks - MediaPipe Tasks: cross-platform APIs and libraries for deploying solutions - MediaPipe Framework: low-level component for building efficient on-device ML pipelines - MediaPipe Studio: visualize, test, and benchmark solutions in the browser - Customizable and deployable across multiple platforms

Use cases

MediaPipe Solutions cover a range of tasks including audio processing, decision making, generative AI, retrieval, text analysis, and vision. Example applications from the project include hand tracking, pose prediction, object detection, eye tracking, and intelligent video reframing.

Getting started

See the repository README or the developer documentation site for setup instructions.

License and pricing

Licensed under Apache-2.0. Open source.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/google-ai-edgemediapipe.svg)](https://yunodb.com/tools/google-ai-edgemediapipe?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/google-ai-edgemediapipe?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/google-ai-edgemediapipe.svg" alt="Featured on YunoDB" height="20"></a>