AI / Computer VisionPrior team work
Facial Recognition Attendance System
This system automates employee attendance using real-time facial recognition. It detects and identifies faces from live camera feeds, logs check-ins and check-outs automatically, and generates attendance reports — removing the friction and inaccuracy of manual tracking, cards, or fingerprint scanners.
PythonOpenCVDeepFaceFastAPI
The Challenge
- Manual and card-based attendance was slow, easy to game (buddy punching), and hard to audit.
- Faces had to be recognized reliably from live feeds under varying lighting and angles.
- The system needed to log check-ins and check-outs in real time without bottlenecks.
- Attendance data had to roll up into reports managers could actually use.
Our Solution
- Built a Python pipeline using OpenCV for face detection and DeepFace for identity recognition.
- Processed live camera frames in real time, matching detected faces against the employee database.
- Automatically logged check-in and check-out events with timestamps as employees appear on camera.
- Exposed the system through a FastAPI service and generated structured attendance reports.
Key Features
Real-Time Recognition
Detects and identifies employees directly from live camera feeds.
Automatic Logging
Check-ins and check-outs recorded the moment a face is recognized.
Attendance Reports
Aggregated, exportable reports for HR and management.
No Touch, No Cards
Eliminates fingerprint scanners and badges — and buddy punching.
What It Does
Real-time
Face-based check-in/out
Contactless
No cards or fingerprints
Auto
Reports without manual entry