MediaPipe for Unity Web | AI, Face Tracking, Computer Vision & AR


🚀 Live Web Demo

💬 Forum Discussion


Integrate MediaPipe Solutions Tasks for Web directly into Unity and bring real-time machine learning to your Web projects. Access powerful AI capabilities across computer vision, audio processing, and text analysis — all running directly in the browser.


Build advanced augmented reality effects, gesture-driven interactions, face and body tracking systems, intelligent object detection pipelines, and language-aware features inside Unity Web.



Designed for Web

Built specifically for Unity Web, the plugin bridges Unity and the MediaPipe Web runtime efficiently and cleanly.


Official API Structure

The architecture mirrors the official MediaPipe Solutions Tasks JavaScript API, making the integration largely self-documenting. Developers can leverage existing documentation, examples, and established patterns to accelerate development.


High Performance & Low Overhead

Built with performance and memory efficiency as core priorities, the plugin is optimized for real-time browser-based applications. It delivers fast execution and remains lightweight enough for mobile web deployments.


Broad Task Support

The plugin currently supports a comprehensive range of MediaPipe Solutions Tasks for Web across three primary domains:


🔊 Audio


📝 Text


👁 Vision


Custom Model Support

Developers can integrate their own machine learning models trained with tools such as MediaPipe Model Maker or TensorFlow. This enables project-specific ML workflows such as custom object detection, domain-specific image classification, specialized audio analysis, and custom text processing directly in the browser.


What Can You Build?


• Browser-based AR face filters and visual effects powered by face detection and landmark tracking


• Gesture-controlled interfaces using real-time hand landmark detection and gesture recognition


• Pose-driven character animation and interactive avatar systems


• Object detection overlays and intelligent scene-aware interactions


• Background removal and segmentation-based visual effects


• Image and text classification workflows running entirely client-side


• Audio classification systems for browser-based applications


• Language detection and embedding-powered text features



Asset uses MediaPipe under Apache 2.0 License; see Third-Party Notices.txt file in package for details.



🧩 MORE ASSETS FOR UNITY WEB


Take your Unity Web projects even further with these powerful companion plugins — all built with the same focus on performance, clean API design, and seamless WebGL integration.


📷 Augmented Reality WebGL — Image Tracking WebAR Bring Image AR directly to the browser — no app download required. Anchor 3D content to real-world image targets with fast, accurate natural-feature tracking. Supports simultaneous multi-image tracking and works seamlessly across Chrome, Safari, Edge, and Firefox on both desktop and mobile. → View on Asset Store


🎥 Recorder WebGL The only Unity plugin built specifically to record gameplay video in WebGL builds. Capture in-game audio, microphone input, or both — with configurable framerate, pause/resume support, and export to MP4 or WebM. Everything runs client-side, no server required. → View on Asset Store


🔥 WebGL API for Firebase Unlock the full Firebase suite inside Unity WebGL — Authentication, Firestore, Realtime Database, Storage, Analytics, Messaging, App Check, Performance, and more. The official Firebase Unity SDK doesn't support WebGL; this plugin fills that gap with a clean, comprehensive integration. → View on Asset Store