flutter_liveness_plugin 0.0.11

Android Only plugin that uses Google ML Kit to determine if the person in the view port is human or not.

Flutter Liveness Plugin

flutter_liveness_plugin is a Flutter plugin built specifically for Android, designed to perform real-time liveness detection using Google ML Kit and the device camera. This plugin is ideal for scenarios requiring secure identity verification, such as KYC processes, biometric login, or authentication systems.

⚠️ iOS support is currently under development.

---

🛠 Platform Support

Platform	Supported
Android	✅ Yes
iOS	🚧 In Progress
Web	❌ No
Desktop	❌ No

---

🚀 Installation

Add the plugin to your Flutter project:

dependencies:
  flutter_liveness_plugin: ^0.0.8

Then run:

flutter pub get

---

🔌 Plugin Capabilities

This plugin performs real-time liveness detection using the front-facing camera, leveraging ML Kit's face detection API.

✅ Features:

• Blink detection
• Smile detection
• On-device processing
• Returns path to captured image (if detection is successful)

---

📸 Basic Usage

1. Import the package

import 'package:flutter_liveness_plugin/flutter_liveness_plugin.dart';

2. Start liveness detection

String? imagePath = (await FlutterLivenessPlugin.captureLiveImage());

if (imagePath != null && imagePath.isNotEmpty) {
  print("Liveness passed. Image path: $imagePath");
} else {
  print("Liveness check failed or was cancelled.");
}

🔁 What You Get

If liveness validation is successful, the method returns a String path to the captured image.

---

📋 Required Android Permissions

You must add camera permissions to your AndroidManifest.xml:

<uses-permission android:name="android.permission.CAMERA" />

Also ensure you request permissions at runtime using a package like permission_handler.

---

📦 How It Works Internally

• Uses CameraX to open the front-facing camera
• Applies ML Kit’s Face Detection to identify facial features
• Tracks eye movement and smiling patterns
• Validates liveness based on blink and smile conditions
• Captures and stores the validated image locally
• Returns the image file path via MethodChannel

---

🧪 Example

The plugin includes a working example app located in the example/ directory. To try it out:

cd example
flutter run

---

🔒 Privacy and Security

• All face processing is done entirely on-device.
• No internet access is used for detection.
• No images or video frames are uploaded or shared.
• The captured image is stored locally and accessible to your app.

---

📅 Roadmap

• ✅ Android support
• ❌ iOS support (in progress)
• ❌ More advanced spoofing checks (e.g. depth, light reflection)
• ❌ Configurable detection thresholds

---

🙌 Contributing

We welcome contributions! Feel free to:

• Submit a pull request
• File an issue
• Suggest enhancements

---

📃 License

This project is licensed under the MIT License.

---

Made with ❤️ by Sahil K.