Skip to main content
FaceNet Android is an on-device face recognition app that performs face recognition by comparing FaceNet embeddings against a vector database. This guide will help you set up and configure the app.

Installation

Download the latest APK from the GitHub Releases page. Alternatively, you can build the app from source:
1

Clone the repository

git clone --depth=1 https://github.com/shubham0204/OnDevice-Face-Recognition-Android
2

Open in Android Studio

Open the project in Android Studio and perform a Gradle sync.
3

Run the app

Connect your Android device or start an emulator, then run the application.

Permissions

When you first launch the app, you’ll need to grant camera permissions. The app cannot function without camera access as it performs real-time face recognition on the camera feed.
Camera permissions are required for the app to work. Without them, face recognition cannot be performed.

Next steps

Once you’ve installed the app and granted permissions:
  1. Add faces to the database to create your recognition dataset
  2. Start recognizing faces in real-time using the camera
  3. Configure advanced options to optimize performance

Build docs developers (and LLMs) love