Skip to content

README.md

Latest commit

 

History

History
232 lines (174 loc) · 6.73 KB

README.md

File metadata and controls

232 lines (174 loc) · 6.73 KB

PerchEye Multiplatform SDK

PerchEye SDK provides advanced facial recognition functionality for mobile applications across multiple platforms, enabling face detection, enrollment, verification, and comparison using embedded TensorFlow Lite models.

🚀 Supported Platforms

  • Android - Native Android SDK with Kotlin/Java support
  • iOS - Native iOS SDK with Swift/Objective-C support
  • Flutter - Dart plugin for cross-platform Flutter applications
  • React Native - JavaScript/TypeScript module for React Native apps

📦 Platform-Specific Documentation

Android SDK

  • Location: /android-aar/
  • Language: Kotlin/Java
  • Integration: Gradle dependency
  • Documentation: Native Android implementation

iOS SDK

  • Location: /ios/
  • Language: Swift/Objective-C
  • Integration: CocoaPods/Swift Package Manager
  • Documentation: Native iOS implementation

Flutter Plugin

  • Location: /flutter/
  • Language: Dart
  • Integration: pub.dev dependency
  • Documentation: Cross-platform Flutter plugin

React Native Module

  • Location: /react-native/
  • Language: JavaScript/TypeScript
  • Integration: npm package
  • Documentation: Cross-platform React Native module

🔧 Quick Start Examples

Android (Kotlin)

val perchEye = PerchEye(context)
perchEye.init()
perchEye.openTransaction()

val result = perchEye.addImage(bitmap)
if (result == ImageResult.SUCCESS) {
    val hash = perchEye.enroll()
    val similarity = perchEye.verify(hash)
}

perchEye.destroy()

iOS (Swift)

let perchEye = PerchEyeSwift()
perchEye.openTransaction()

let result = perchEye.load(image: uiImage)
if result == .success {
    let hash = perchEye.enroll()
    let similarity = perchEye.verify(hash: hash)
}

perchEye.destroy()

Flutter (Dart)

await PerchEye.init();
await PerchEye.openTransaction();

final result = await PerchEye.addImage(base64Image);
if (result == 'SUCCESS') {
    final hash = await PerchEye.enroll();
    final similarity = await PerchEye.verify(hash);
}

await PerchEye.destroy();

React Native (JavaScript)

import { openTransaction, addImage, enroll, verify } from 'react-native-perch-eye';

await openTransaction();
const result = await addImage(base64Image);
if (result === 'SUCCESS') {
    const hash = await enroll();
    const similarity = await verify(hash);
}

🎯 Core Features

Universal Functionality

All platform implementations provide:

  • Face Detection - Identify human faces in images
  • Face Enrollment - Generate unique biometric hashes
  • Face Verification - Compare faces against stored hashes
  • Batch Processing - Handle multiple images efficiently
  • Offline Operation - No internet connection required

ImageResult Status Codes

All platforms return consistent status codes:

  • SUCCESS - Operation completed successfully
  • FACE_NOT_FOUND - No face detected in the image
  • FILE_NOT_FOUND - Image file not found (Android only)
  • TRANSACTION_NOT_OPEN - No active transaction
  • SDK_NOT_INITIALIZED - SDK not properly initialized
  • INTERNAL_ERROR - Internal processing error

🔐 Security & Privacy

  • Offline Processing - All computation happens on-device
  • No Data Transmission - Biometric data never leaves the device
  • Hash-Based Storage - Only mathematical representations are stored
  • Privacy Compliant - Meets GDPR and privacy regulations

📱 Demo Applications

Each platform includes a fully functional demo application:

🛠 Development Setup

Prerequisites

  • Android: Android Studio, Gradle, API level 24+
  • iOS: Xcode 12+, iOS 14.0+, Swift 5.0+
  • Flutter: Flutter SDK 3.0+, Dart 2.17+
  • React Native: Node.js 18+, React Native 0.70+

Building from Source

# Clone the repository
git clone https://github.com/Onix-Systems/PerchEye-SDK-Multiplatform.git

# Build Android SDK
cd android && ./gradlew build

# Build iOS SDK  
cd ios && xcodebuild -project PerchEyeFramework.xcodeproj

# Build Flutter plugin
cd flutter && flutter packages get

# Build React Native module
cd react-native && npm install

📊 Performance Characteristics

Processing Speed

  • Face Detection: ~50-100ms per image
  • Hash Generation: ~100-200ms per face
  • Verification: ~10-50ms per comparison
  • Multi-threading: Supported on all platforms

Memory Usage

  • Runtime Memory: ~50-100MB active usage
  • Model Size: ~10-20MB embedded model
  • Hash Size: ~2-5KB per face encoding

Accuracy Metrics

  • Detection Rate: >95% for clear frontal faces
  • False Accept Rate: <0.1% at 0.8 threshold
  • False Reject Rate: <5% at 0.8 threshold

🔧 Integration Guides

Platform-Specific Setup

Android

dependencies {
    implementation(files("libs/perch-eye-1.0.3-4.aar"))
}

iOS

# Download PerchEye Framework from the official source
# Drag PerchEyeFramework.xcframework into your Xcode project
# Ensure the framework is added to Frameworks, Libraries, and Embedded Content
# Set Embed & Sign for the framework

Flutter

# Run this in your Flutter project:

flutter pub add perch_eye

React Native

npm install react-native-perch-eye

🌐 Web Documentation

Visit our comprehensive web documentation at:

📞 Support & Resources

📄 License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

🤝 Contributing

Contributions are welcome! Please read our contributing guidelines and submit pull requests for any improvements.

Note: Each platform maintains its own specific documentation and examples. Refer to the individual platform folders for detailed implementation guides and platform-specific features.