https://github.com/tharushaudana/flutter_face_auth_2
This Flutter project implements face authentication using the FaceNet512 model, storing face data (as Float32 arrays) and names in Firebase Firestore. It allows users to capture and store their face data for subsequent identification, predicting identities based on cosine similarity.
https://github.com/tharushaudana/flutter_face_auth_2
face-authentication face-detection face-recognition facenet firebase machine-learning tensorflow-lite
Last synced: about 1 month ago
JSON representation
This Flutter project implements face authentication using the FaceNet512 model, storing face data (as Float32 arrays) and names in Firebase Firestore. It allows users to capture and store their face data for subsequent identification, predicting identities based on cosine similarity.
- Host: GitHub
- URL: https://github.com/tharushaudana/flutter_face_auth_2
- Owner: tharushaudana
- Created: 2025-07-30T20:43:18.000Z (2 months ago)
- Default Branch: main
- Last Pushed: 2025-07-30T21:51:01.000Z (2 months ago)
- Last Synced: 2025-07-30T23:00:14.094Z (2 months ago)
- Topics: face-authentication, face-detection, face-recognition, facenet, firebase, machine-learning, tensorflow-lite
- Language: Dart
- Homepage:
- Size: 60.1 MB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Flutter Face Authentication 2
This is the updated version of the Flutter-based face authentication application. It detects faces using **Google ML Kit's face detection** and uses the **FaceNet512** model to recognize and differentiate users. Encoded face data is stored and retrieved from **Firebase Firestore**.
✅ Fully compatible with:
- **Flutter** 3.32.6 • channel stable • [flutter.git](https://github.com/flutter/flutter.git)
- **Framework** revision 077b4a4ce1 (2025-07-08)
- **Engine** revision 72f2b18bb0
- **Dart** 3.8.1 • DevTools 2.45.1
---
## Features
- 👁️ **Face Detection with Google ML Kit**: Uses `google_mlkit_face_detection` to detect faces in real time.
- 🧠 **Face Encoding with FaceNet512**: Converts detected faces into 512-length feature vectors using the FaceNet512 model.
- ☁️ **Cloud Firestore Integration**: Stores and retrieves encoded face data (Float32 arrays) and associated names.
- 🔄 **Real-Time Face Authentication**: Predicts user identity by comparing current face data with stored vectors using cosine similarity.
- 📦 **On-device Inference with TFLite**: Utilizes `tflite_flutter` for running the FaceNet model locally on the device.
## Dependencies
```yaml
cupertino_icons: ^1.0.2
camera: ^0.10.5+5
google_mlkit_face_detection: ^0.13.1
image: ^3.0.2
tflite_flutter: ^0.11.0
cloud_firestore: ^4.13.2
firebase_core: ^2.23.0
````
## Screenshots
### App Interface
### Register New Face Page (when a face not in the database is detected)
### Predicted Page
### When a Face Is Not Detected Correctly
### Data Storage in Firestore
## Getting Started
1. Clone this repository:
```bash
git clone https://github.com/tharushaudana/flutter_face_auth_2.git
cd flutter_face_auth_2
```
2. Install dependencies:
```bash
flutter pub get
```
3. Set up Firebase for Android and iOS ([Firebase Setup Guide](https://firebase.google.com/docs/flutter/setup)).
4. Run the app:
```bash
flutter run
```
## Purpose
This project demonstrates mobile face authentication using deep learning and real-time face detection. It is suitable for secure identity verification systems where user faces are encoded, stored, and compared efficiently using on-device AI and cloud storage.