← Back to selected work

COMPUTER VISION / BEHAVIORAL AI / REAL-TIME ANALYTICS

ATTENTION SPAN TRACKER

A real-time Computer Vision and Behavioral AI system that monitors user attention and productivity during computer usage.

Case
01 / 04
Stack
Python · OpenCV
Year
2026
CASE STUDY
OverviewWorkflowImplementationTakeaway

01 / OVERVIEW

Project overview

The Attention Span Tracker is a real-time Computer Vision and Behavioral AI system that monitors user attention and productivity during computer usage.

It processes the webcam feed to detect faces, then analyzes gaze and posture alongside the contextual task being performed to compute an attention score. The results flow into a browser-based dashboard with real-time analytics and productivity insights.

All processing happens locally on the machine, keeping video and behavioral data private.

02 / WORKFLOW

System workflow

  1. WEBCAM01
  2. FACE DETECTION02
  3. GAZE ANALYSIS03
  4. POSTURE ASSESSMENT04
  5. CONTEXTUAL TASK CLASSIFICATION05
  6. ATTENTION SCORING06
  7. REAL-TIME DASHBOARD07

03 / PIPELINE

Computer vision pipeline

  • Webcam-based face detection
  • Gaze analysis
  • Posture assessment
  • Contextual task classification

04 / ANALYSIS

Attention analysis

  • Attention scoring
  • Real-time analytics
  • Productivity insights
  • Visualization

05 / DASHBOARD

Dashboard

A browser-based dashboard presents attention scoring, real-time analytics and productivity insights as they happen.

06 / STACK

Technology stack

  • Python
  • OpenCV
  • MediaPipe
  • HTML
  • CSS
  • JavaScript

07 / PRIVACY

Privacy / local processing

LOCAL PROCESSING

All processing happens locally on the machine. Video and behavioral data are not sent anywhere — the analysis runs entirely on-device.

PROJECT TAKEAWAY

What it adds up to.

The Attention Span Tracker shows how computer vision, behavioral analysis and real-time analytics can be combined into one privacy-preserving pipeline — from raw webcam input to an actionable browser-based dashboard.