Problem

Manual attendance tracking is slow, error-prone, and easily manipulated. Educational and organizational institutions need reliable, automated solutions.

Solution

Ahmed and his co-developer built a complete pipeline — from live camera input to face detection (YOLOv11), identity matching (FaceNet), and backend storage — with a clean REST API layer.

Architecture

The system processes live video through a Python-based CV pipeline: YOLOv11 detects faces in each frame, FaceNet generates embeddings for identity matching against registered profiles, and attendance records are persisted in MySQL via a Node.js REST API.

Key Features

Real-time video processing pipeline with live camera integration
YOLOv11-powered face detection with high accuracy
FaceNet-based identity matching against registered profiles
MySQL-backed attendance record management
RESTful API for attendance query and reporting

Challenges

Optimizing inference latency for real-time video; handling lighting variability and partial occlusion in face detection.

Outcomes

A working automated attendance system with real-world applicability for classrooms, offices, or events.