Computer Vision Project

Face Recognition Attendance System

A vision-based attendance system that recognises enrolled students from a camera feed and records attendance automatically.

Intermediate7–9 weeksCSE · IT · MCAComputer VisionAI & Machine Learning

Project overview

Students are enrolled once by capturing a set of face images. During a class the camera detects faces in the frame, matches each against the enrolled embeddings, and marks attendance in the database with a timestamp, without any manual roll call.

It is one of the most requested computer vision projects because the demo is immediate. The engineering that earns marks is in the enrolment quality, the matching threshold and the anti-spoofing discussion.

Problem statement

Manual roll call consumes teaching time in every period and is easy to manipulate through proxy attendance. Card-based systems solve the time cost but not the proxy problem, since a card can be handed over.

Proposed solution

Use a face detector to locate faces, generate embeddings with a pretrained recognition model, and match against enrolled embeddings using a distance threshold tuned on validation data. Record attendance once per student per session, with an unknown-face path that never silently marks the wrong person.

Key features

  • Guided enrolment capturing multiple angles per student
  • Real-time face detection from a webcam or IP camera
  • Embedding-based recognition with a tuned distance threshold
  • Duplicate suppression so one student is marked once per session
  • Unknown face handling with a review queue rather than a forced match
  • Basic liveness check to reject a photograph held to the camera
  • Attendance dashboard with per-subject and per-student reports
  • CSV export for the department register

Technology stack

Vision

PythonOpenCVface_recognition / dlibMediaPipe

Models

FaceNet-style embeddingsHaar cascade or MTCNN detector

Backend

FlaskMySQL or PostgreSQL

Front end

React or Flask templatesChart.js

System architecture

The flow below is what goes into your architecture diagram and the corresponding chapter of the report.

  1. Enrolment captures images, detects the face region and stores the averaged embedding
  2. Capture loop reads frames from the camera at a controlled rate
  3. Detection stage locates all faces in the frame
  4. Recognition stage computes embeddings and finds the nearest enrolled match
  5. Decision stage applies the distance threshold and the liveness check
  6. Attendance service writes the record and prevents duplicates within a session

Expected output

  • Live recognition view with names and confidence overlaid
  • Attendance records with timestamps per session
  • Recognition accuracy measured on a held-out set of images
  • Subject-wise attendance percentage report

Suitable branches, difficulty and duration

Suitable branches

Difficulty level

Intermediate. Comfortable for a team where at least one member has worked in this stack before.

Development duration

7–9 weeks including documentation, tested and ready to demonstrate. Milestones are planned around your review dates.

Documentation, review, demo and viva support

Documentation support

Abstract, problem statement, literature survey guidance, system architecture, ER and UML diagrams, module descriptions, test cases and screenshots — prepared in your department format.

Project review support

Delivery is split across your review calendar so review one has a design and plan, review two has working modules, and the final review has the integrated, tested system.

Demo support

A rehearsed demo path, seeded sample data, a reset script, and a recorded walkthrough kept as a backup in case something fails in the review room.

Viva preparation

A written question bank for this domain, a code walkthrough session, and practice at explaining each module in your own words — because the examiner will ask you.

Customisation options

Colleges reject duplicate topics, so most projects are adjusted before we start. These are the usual directions for this one.

  • Add mask-tolerant recognition using a periocular model
  • Move inference to a Raspberry Pi at the classroom door
  • Add automatic notification to parents on absence
  • Combine with the campus event portal for event check-in
Talk to CodeLab69

Enquire about the Face Recognition Attendance System project

Send your branch, college and first review date. You will get confirmation of scope, what the customisation would look like, and a delivery plan mapped to your reviews.