Z-ENG: Computer Vision and Intelligent Visitor Analytics for Smart Events

2026-2027 ősz

Nincs megadva

Téma leírása

Description

This topic focuses on extending the computer vision capabilities of the event platform to better understand visitor behavior and improve personalization. Students will develop AI models that analyze visitor interactions using cameras while respecting privacy considerations. The developed modules should improve audience analytics, multimedia recommendation, and real-time event monitoring.

Possible Subtopics

2.1 Visitor Behavior Analysis

Develop computer vision algorithms that estimate visitor attention, engagement, and interaction with exhibition stands.

Tasks

  • Detect visitors.
  • Estimate dwell time.
  • Analyze viewing behavior.
  • Estimate attention levels.
  • Evaluate performance on real event scenarios.

2.2 Multi-Camera Visitor Tracking

Develop algorithms for tracking visitors across multiple cameras.

Tasks

  • Implement person re-identification.
  • Associate identities between cameras.
  • Visualize visitor movement.
  • Generate movement statistics.

2.3 Privacy-Preserving Analytics

Investigate techniques that enable visitor analytics without storing personally identifiable information.

Tasks

  • Study GDPR-compliant approaches.
  • Design anonymous visitor representations.
  • Compare privacy-preserving techniques.
  • Evaluate utility versus privacy.

2.4 Intelligent Audience Analytics Dashboard

Develop visualization tools for organizers to monitor visitor activity in real time.

Tasks

  • Design dashboards.
  • Display visitor statistics.
  • Generate heat maps.
  • Visualize engagement metrics.
  • Produce automatic reports.

Required Skills

  • Python
  • OpenCV
  • Deep Learning
  • PyTorch or TensorFlow
  • Computer Vision
  • Data Visualization

Maximális létszám: 4 fő