🛡️ Face Vault - Biometric Password Manager

Face Vault is a next-generation desktop application for securely managing your passwords, protected by your own identity. It combines modern web technologies with advanced AI-powered facial recognition to ensure that your passwords remain yours alone.

✨ Key Features

• ⚡ Biometric Security: Log in using your face with professional-grade recognition models.
• 🛡️ 3D Liveness Detection (Anti-Spoofing): Advanced models to prevent photo/video-based spoofing attempts.
• 🚀 GPU Acceleration: Fully optimized for Windows using DirectML, leveraging any GPU (NVIDIA, AMD, Intel) for lightning-fast scanning.
• 📂 Secure Local Storage: Your data stays on your machine, encrypted in a local SQLite database.
• 🎭 Dual Authentication: Seamlessly switch between Face ID and a traditional Master Password.
• 🌈 Modern UI/UX: Sleek, glassmorphic design built with React and Framer Motion elements.

---

🏗️ Technology Stack

Frontend & Application Shell

• Electron: Cross-platform desktop framework.
• React (Vite 8.0): Ultra-fast frontend framework.
• TypeScript: Type-safe logic for both main and renderer processes.
• Lucide Icons: Premium, consistent iconography.

AI & Biometrics Engine (Python)

• ONNX Runtime (DirectML): Execution engine for AI models on Windows GPUs.
• InsightFace: Industry-standard library for face analysis.
• PyTorch: Backing tensor computations.
• MiniFASNetV2: Specialized light-weight model for anti-spoofing detection.

Data Management

• better-sqlite3: High-performance, synchronous SQLite driver for Node.js.

---

🧬 AI Models & Workflow

Face Vault uses a multi-stage AI pipeline to ensure both accuracy and security:

1. Face Detection: Identifies facial coordinates in the live camera feed.
2. Liveness Check: Analyzes texture and depth patterns to ensure the person is real (not a photo/mask).
3. Feature Extraction: Generates a unique "Biometric Signature" from the face.
4. Identity Matching: Compares the live signature against the encrypted local database using cosine similarity.

Workflow Diagram

graph TD
    A[Camera Feed] --> B[Face Detection]
    B --> C{Liveness Detection}
    C -- Real --> D[Feature Extraction]
    C -- Spoof --> E[Access Denied]
    D --> F[Database Comparison]
    F -- Match Found --> G[Unlock Vault]
    F -- No Match --> H[Identity Error]

Loading

---

🛠️ Project Structure

• /electron-app: The core application codebase.
  • /src: React frontend and component library.
  • /electron: Main process and Preload scripts for bridge security.
  • /python: Bundled, portable Python environment for the production build.
• /Face Recognition: Python scripts handling AI inference and biometric storage.
  • /models: Directory for pre-trained AI weights (ONNX format).
  • /face_db: Encrypted biometric templates for registered users.
• /release: (Auto-generated) Contains the final Windows Installer (.exe).

---

🚀 Getting Started

Prerequisites

• Node.js (v18+)
• Python 3.11+
• A Windows 10/11 device with a Webcam.

Development Setup

1. Clone the Repo:

   git clone [repository-url]
   cd "Password Manager Using Face Recognition/electron-app"

2. Install Node Dependencies:

   npm install

3. Configure Python Environment: Create a venv and install requirements:

   pip install -r requirements-bundle.txt

4. Run in Dev Mode:

   npm run dev

Building the Production App

To create the final Windows Installer:

npm run electron:build

The installer will be generated in electron-app/release/Face Vault Setup [version].exe.

---

🔒 Privacy & Security

• Local-First Architecture: Your biometrics and passwords never leave your computer. We do not use cloud storage.
• Encryption: Passwords are encrypted before storage using system-level salts and a custom encryption layer.
• Isolation: The AI backend runs in a separate process with limited permissions to increase security.

---

Developed by Makious Face Vault - Your Identity is the Key.