Grabpic Intelligent Identity & Retrieval Engine

A high-performance image processing backend designed for large-scale events. It uses facial recognition to automatically group images and provides a secure "Selfie-as-a-Key" retrieval system.

🏗️ Architecture Design & Schema

Tech Stack: Python, Flask, SQLite, face_recognition

System Flow:

1. Discovery & Transformation (Ingestion Engine): Scans a local storage folder (raw_images), extracts mathematical facial encodings using C++ models, and persists them into a relational database.
2. Web API: A lightweight Flask server handling selfie authentication and search token retrieval.

Relational Database Schema: The database maps one image to potentially many users.

• Images Table: image_id (PK), file_path
• Faces Table: grab_id (PK), face_encoding (BLOB)
• Image_Faces Table: image_id (FK), grab_id (FK) -> Many-to-Many mapping.

🚀 Setup & Installation

1. Create and activate a virtual environment:

   python3 -m venv venv
   source venv/bin/activate
   

2.Install dependencies (requires specific setuptools version for model compatibility):

pip install flask face_recognition numpy setuptools==76.0.0

⚙️ Running the Engine
Step 1: Run the Ingestion Engine (Ensure you have a raw_images folder with photos)
```bash
python3 ingest.py

Step 2: Start the Web API
```bash
python3 app.py
📖 API Documentation & cURL Tests
1. Selfie Authentication
Authenticates a user via an image file and returns their authorizer grab_id.

Endpoint: POST /authenticate

cURL Command:
```bash
curl -X POST -F "file=@test_selfie.jpg" [http://127.0.0.1:8000/authenticate](http://127.0.0.1:8000/authenticate)

2. Data Extraction
Fetches all indexed images for a specific user.

Endpoint: GET /images/<grab_id>

cURL Command:

curl -X GET [http://127.0.0.1:8000/images/1](http://127.0.0.1:8000/images/1)