English | 简体中文

A Flutter plugin for FaceAISDK offline face recognition. Supports Platform View (Built-in UI mode) and direct API calls.

Features

• Offline Face Recognition: High-performance offline face identification and verification.
• Liveness Detection: Supports Action, Silent, and Color Flash liveness detection.
• 1:1 Face Verification: Compare two faces for similarity.
• Platform View Support: Seamlessly integrate the camera preview and UI into your Flutter widget tree.
• Face Management: Register (via Camera or Image), delete, and check for face features locally.

Getting Started

Android

1. Add the following to your AndroidManifest.xml:

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

2. Ensure minSdkVersion is at least 21.

iOS

1. Add the following to your Info.plist:

   <key>NSCameraUsageDescription</key>
   <string>FaceAISDK needs camera access for face enrollment and liveness verification.</string>
   

2. Ensure the deployment target is iOS 15.5 or higher.

Usage

1. Platform View (Built-in UI)

Use FaceRecognitionView to show a camera preview with built-in face recognition UI.

import 'package:FaceRecognition_Flutter/FaceRecognition_Flutter.dart';

FaceRecognitionView(
  needShowConfirmDialog: true, // Default is true
  onViewCreated: (FaceRecognitionController controller) {
    // You can use the controller to start or stop scanning
    // controller.startScan();
  },
  creationParams: {
    'threshold': 0.84,
  },
)

2. Direct API Calls

The FaceRecognitionFlutter class provides several static methods for face operations.

Face Verification (1:1 + Liveness)

final result = await FaceRecognitionFlutter.faceVerify(
  faceId: "user_123",
  threshold: 0.84,
  livenessType: 1, // 1: Action, 2: Action+Color, 3: Color, 4: Silent
  motionLivenessTypes: "1,2,3,4,5", // 1: Mouth, 2: Smile, 3: Blink, 4: Shake, 5: Nod
);

if (result.code == FaceRecognitionResultCode.verifySuccess) {
  print("Face verified successfully. Similarity: ${result.similarity}");
}

Face Registration (via Camera)

final result = await FaceRecognitionFlutter.addFaceBySDKCamera(
  faceId: "new_user_456",
  addFacePerformanceMode: 1, // 1: Fast, 2: Precise
  needShowConfirmDialog: true, // Show confirmation dialog after capture
);

Face Registration (via Image Base64)

final result = await FaceRecognitionFlutter.addFaceBySDKImage(
  faceId: "user_123",
  imageBase64: "data:image/jpeg;base64,...",
);

Face Management

// Check if face exists
bool exists = await FaceRecognitionFlutter.isFaceExist("user_123");

// Get face feature (1024-bit string)
final result = await FaceRecognitionFlutter.getFaceFeature("user_123");

// Delete face
await FaceRecognitionFlutter.deleteFaceFeature("user_123");

Result Codes

Use FaceRecognitionResultCode to handle outcomes:

• verifySuccess (1): Verification passed.
• verifyFailed (2): Verification failed.
• motionLivenessSuccess (3): Action liveness passed.
• allLivenessSuccess (10): All liveness checks passed.
• cancel (0): User cancelled.

Other SDK API Demo

iOS SDK： https://github.com/FaceAISDK/FaceAISDK_iOS
Android： https://github.com/FaceAISDK/FaceAISDK_Android
uniApp UTS https://github.com/FaceAISDK/FaceAISDK_uniapp_UTS Flutter： https://github.com/FaceAISDK/FaceRecognition_Flutter React Native https://github.com/FaceAISDK/FaceRecognition_ReactNative

---

FaceRecognition_Flutter (简体中文)

适用于 FaceAISDK 离线人脸识别的 Flutter 插件。支持 Platform View（内置 UI 模式）和直接 API 调用。

特性

• 离线人脸识别: 高性能离线人脸比对与识别。
• 活体检测: 支持动作活体、静默活体及炫彩活体检测。
• 1:1 人脸比对: 比较两张人脸的相似度。
• 支持 Platform View: 将相机预览和 UI 无缝集成到 Flutter 组件树中。
• 人脸管理: 本地注册（相机或图片）、删除和检查人脸特征值（1024 位）。

入门指南

Android

1. 在 AndroidManifest.xml 中添加：

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

2. 确保 minSdkVersion 至少为 21。

iOS

1. 在 Info.plist 中添加：

   <key>NSCameraUsageDescription</key>
   <string>FaceAISDK needs camera access for face enrollment and liveness verification.</string>
   

2. 确保 Deployment Target 至少为 iOS 15.5。

使用示例

1. Platform View (内置 UI 模式)

使用 FaceRecognitionView 组件在 Flutter 中嵌入带有引导 UI 的相机界面。

import 'package:FaceRecognition_Flutter/FaceRecognition_Flutter.dart';

FaceRecognitionView(
  needShowConfirmDialog: true, // 录入成功后是否显示确认弹窗，默认为 true
  onViewCreated: (FaceRecognitionController controller) {
    // 可以通过控制器手动开始或停止扫描
    // controller.startScan();
  },
  creationParams: {
    'threshold': 0.84,
  },
)

2. API 调用模式

通过 FaceRecognitionFlutter 类直接调用功能接口。

人脸核验 (1:1 + 活体检测)

final result = await FaceRecognitionFlutter.faceVerify(
  faceId: "user_123",
  threshold: 0.84,
  livenessType: 1, // 1: 动作, 2: 动作+炫彩, 3: 炫彩, 4: 静默
  motionLivenessTypes: "1,2,3,4,5", // 1: 张嘴, 2: 微笑, 3: 眨眼, 4: 摇头, 5: 点头
);

if (result.code == FaceRecognitionResultCode.verifySuccess) {
  print("人脸核验成功，相似度: ${result.similarity}");
}

人脸采集注册 (通过图片 Base64)

final result = await FaceRecognitionFlutter.addFaceBySDKImage(
  faceId: "user_123",
  imageBase64: "data:image/jpeg;base64,...",
);

人脸管理

// 检查人脸是否存在
bool exists = await FaceRecognitionFlutter.isFaceExist("user_123");

// 获取 1024 位人脸特征字符串
final result = await FaceRecognitionFlutter.getFaceFeature("user_123");

// 删除人脸信息
await FaceRecognitionFlutter.deleteFaceFeature("user_123");

结果状态码

使用 FaceRecognitionResultCode 处理不同的返回结果：

• verifySuccess (1): 核验成功。
• verifyFailed (2): 核验失败。
• motionLivenessSuccess (3): 动作活体通过。
• allLivenessSuccess (10): 所有活体检测通过。
• cancel (0): 用户取消操作。

Other SDK API Demo

iOS SDK： https://github.com/FaceAISDK/FaceAISDK_iOS
Android： https://github.com/FaceAISDK/FaceAISDK_Android
uniApp： https://github.com/FaceAISDK/FaceAISDK_uniapp_UTS
Flutter： https://github.com/FaceAISDK/FaceRecognition_Flutter
React Native https://github.com/FaceAISDK/FaceRecognition_ReactNative