FaceCog Liveness Verification

An Angular workspace containing a liveness verification library and a showcase demo application.

Project Structure

facecog-ionic-front/
├── projects/
│   ├── facecog-liveness-verification/        # @naniteninja/liveness-verification
│   │   ├── src/lib/
│   │   │   ├── models/
│   │   │   │   ├── types/                    # Type definitions (one per file)
│   │   │   │   │   ├── liveness-backend.type.ts
│   │   │   │   │   ├── verification-flow-step.type.ts
│   │   │   │   │   ├── liveness-action.type.ts
│   │   │   │   │   ├── pose-category.type.ts
│   │   │   │   │   ├── pose-difficulty.type.ts
│   │   │   │   │   ├── detection-strategy.type.ts
│   │   │   │   │   └── index.ts
│   │   │   │   │
│   │   │   │   ├── interfaces/               # Interface definitions (one per file)
│   │   │   │   │   ├── liveness-verification-config.interface.ts
│   │   │   │   │   ├── liveness-result.interface.ts
│   │   │   │   │   ├── pose-definition.interface.ts
│   │   │   │   │   ├── ... (13 more interfaces)
│   │   │   │   │   └── index.ts
│   │   │   │   │
│   │   │   │   ├── constants/                # Constant values
│   │   │   │   │   ├── default-liveness-config.constant.ts
│   │   │   │   │   ├── pose-definitions.constant.ts
│   │   │   │   │   └── index.ts
│   │   │   │   │
│   │   │   │   ├── utils/                    # Utility functions
│   │   │   │   │   ├── pose.utils.ts
│   │   │   │   │   └── index.ts
│   │   │   │   │
│   │   │   │   └── index.ts
│   │   │   │
│   │   │   ├── services/                     # Core services
│   │   │   │   ├── backends/
│   │   │   │   │   ├── backend-adapter.interface.ts
│   │   │   │   │   └── mock-backend.service.ts
│   │   │   │   ├── camera.service.ts
│   │   │   │   ├── liveness-config.service.ts
│   │   │   │   ├── pose-selection.service.ts
│   │   │   │   └── video-recorder.service.ts
│   │   │   │
│   │   │   ├── liveness-verification.component.ts
│   │   │   └── liveness-verification.module.ts
│   │   │
│   │   ├── src/public-api.ts
│   │   ├── package.json
│   │   ├── ng-package.json
│   │   └── README.md
│   │
│   └── facecog-liveness-verification-test/   # Showcase Demo App
│       └── src/
│           ├── app/
│           │   ├── liveness-verification/    # Full implementation
│           │   ├── verification-demo/        # FaceTec demo
│           │   └── home/                     # Home page
│           ├── assets/
│           ├── environments/
│           └── theme/
│
├── dist/
│   └── naniteninjas-liveness-verification/   # Built library
│
├── angular.json
├── tsconfig.json
└── package.json

Quick Start

Install Dependencies

npm install

Development Server

npm start

Starts facecog-liveness-verification-test on http://localhost:4200.

Build Library

npm run build:lib

Outputs to dist/naniteninjas-liveness-verification/.

Build Demo App

npm run build:prod

Outputs to www/.

Library Usage

Local Development (No Build Required)

The workspace is configured with TypeScript path mapping:

// tsconfig.json
{
  "paths": {
    "@naniteninja/liveness-verification": [
      "projects/facecog-liveness-verification/src/public-api.ts"
    ]
  }
}

Import Examples

// Types (single export per file)
import { LivenessBackend } from '@naniteninja/liveness-verification';
import { VerificationFlowStep } from '@naniteninja/liveness-verification';
import { PoseCategory } from '@naniteninja/liveness-verification';

// Interfaces (single export per file)
import { LivenessVerificationConfig } from '@naniteninja/liveness-verification';
import { LivenessResult } from '@naniteninja/liveness-verification';
import { PoseDefinition } from '@naniteninja/liveness-verification';

// Constants
import { DEFAULT_LIVENESS_CONFIG } from '@naniteninja/liveness-verification';
import { POSE_DEFINITIONS } from '@naniteninja/liveness-verification';

// Utils
import { getPoseById, getPosesByCategory } from '@naniteninja/liveness-verification';

// Services
import { LivenessConfigService, PoseSelectionService } from '@naniteninja/liveness-verification';

// Module/Component
import { LivenessVerificationModule, provideLivenessVerification } from '@naniteninja/liveness-verification';

Module-based Usage

import { LivenessVerificationModule } from '@naniteninja/liveness-verification';

@NgModule({
  imports: [
    LivenessVerificationModule.forRoot({
      apiUrl: 'https://api.example.com',
      backends: ['amazon', 'facetec']
    })
  ]
})
export class AppModule {}

Standalone Usage

import { provideLivenessVerification } from '@naniteninja/liveness-verification';

bootstrapApplication(AppComponent, {
  providers: [
    provideLivenessVerification({ backends: ['mock'] })
  ]
});

Template

<facecog-liveness-verification
  [pose]="5"
  (verificationComplete)="onComplete($event)"
  (verificationError)="onError($event)">
</facecog-liveness-verification>

Available Scripts

| Script | Description | |--------|-------------| | npm start | Start the showcase demo app | | npm run build | Build the demo app | | npm run build:prod | Build demo app for production | | npm run build:lib | Build the library | | npm run package | Build and pack the library | | npm run npm:pub | Publish library to npm | | npm run bump:lib | Bump library version | | npm run lint | Lint the demo app |

Publishing

# Update version
npm run bump:lib

# Build and pack
npm run package

# Publish
npm run npm:pub

Code Organization Principles

Single Export Per File

All types, interfaces, and enums follow the single-export-per-file pattern:

// ✅ Good: types/liveness-backend.type.ts
export type LivenessBackend = 'amazon' | 'openpose' | 'facetec' | 'azure' | 'mock';

// ✅ Good: interfaces/liveness-result.interface.ts
export interface LivenessResult {
  success: boolean;
  confidence: number;
  // ...
}

Barrel Exports (index.ts)

Each folder has an index.ts that re-exports its contents:

// types/index.ts
export { LivenessBackend } from './liveness-backend.type';
export { VerificationFlowStep } from './verification-flow-step.type';
// ...

// models/index.ts
export * from './types';
export * from './interfaces';
export * from './constants';
export * from './utils';

Features

• 40+ Poses: Face, head, hand, gesture, and combined poses
• Multi-Backend: Amazon, FaceTec, OpenPose, Azure, Mock
• TensorFlow.js: MoveNet for pose detection
• MediaPipe: Face & Hand Landmarker
• Configurable: Thresholds, retries, video recording

Tech Stack

• Angular 20
• Ionic 8
• TensorFlow.js
• MediaPipe
• RxJS

License

MIT