Face Verification System

A facial verification MVP built using InsightFace, Streamlit, OpenCV, and MongoDB Atlas.

Features

• User registration with face detection and metadata
• View, edit, and delete registered users
• Real-time face detection and recognition
• Activity logs with filtering and statistics
• MongoDB Atlas integration for cloud storage

Setup and Installation

Prerequisites

• Python 3.8 or higher
• A MongoDB Atlas account (free tier is sufficient)
• Webcam for real-time recognition

Installation

1. Clone this repository:

   git clone https://github.com/Saifullah3711/facial-recognition-st
   
   cd facial-recognition-st
   

2. Install dependencies:

   pip install -r requirements.txt
   

3. Create a .env file in the project root and add your MongoDB Atlas connection string:

   MONGODB_URI=mongodb+srv://<username>:<password>@<cluster>.mongodb.net/<dbname>?retryWrites=true&w=majority
   

   Replace <username>, <password>, <cluster>, and <dbname> with your actual MongoDB Atlas details.

Windows Installation Notes

If you encounter DLL loading errors with ONNX or InsightFace, try these solutions:

1. RECOMMENDED SOLUTION: Downgrade ONNX to version 1.16.1 (this fixes the "DLL load failed while importing onnx_cpp2py_export" error):

   pip install onnx==1.16.1
   

2. Install the Visual C++ Redistributable:

   • Download and install the Microsoft Visual C++ Redistributable
   • Restart your system
   • Run the application again

3. If issues persist, the application will automatically fall back to using OpenCV's face detection, which is less accurate but more compatible.

Reference: ONNX GitHub Issue #6267

Running the Application

Run the Streamlit app: