Develop a real-time computer vision solution to assess the engagement of the audience during an online session (lecture, conference), by analyzing their faces, emotional response and/or pose based on their webcam video feeds.

Required interest(s)

  • Face Recognition
  • Applied Deep Learning
  • Human-Centered and Community-Minded Information Systems

What do you get

  • A challenging assignment within a practical environment
  • € 1000 compensation, € 500 + lease car or € 600 + living space
  • Professional guidance
  • Courses aimed at your graduation period
  • Support from our academic Research center at your disposal
  • Two vacation days per month

What you will do

  • 65% Research
  • 10% Analyze, design, realize
  • 25% Documentation

Nowadays, video conferencing has become more ubiquitous than ever before. This is especially true after the global lockdown prompted by the COVID-19 pandemic, during which work meetings, university lectures, school lessons, conferences, and even social activities, were moved to the realm of online video calls.

While this has posed several challenges and limitations compared to in-person gatherings, it has also offered many advantages and new possibilities. Concretely, the fact that every participant could stream a video of themselves during such encounters, entails a very powerful data source for video analytics and affective computing.

A key aspect of online sessions like lectures and conferences, and a strong indicator of their quality, is the audience engagement. Until now, engagement has been generally measured via traditional means, such as questionnaires and feedback forms. But these means require participants to be proactive, and they can hardly ensure objectivity. On the contrary, analyzing the video recordings of participants and attendees could provide a passive indicator of engagement, and avoid subjectivity-derived issues such as negativity bias.

Face and pose analysis techniques –such as facial expression or emotion recognition, face action units detection, pose estimation, or eye-gaze estimation– could be used to model human engagement, and potentially applied to automatically detect and rate the level of engagement among an online audience in video sessions. Having such insights could help lecturers and speakers assess the quality of their talks, find out possible weak interest points, discover which content or storytelling resources trigger more attention, and further improve their sessions.

About Info Support Research Center

We anticipate on upcoming and future challenges and ensures our engineers develop cutting-edge solutions based on the latest scientific insights. Our research community proactively tackles emerging technologies. We do this in cooperation with renowned scientists, making sure that research teams are positioned and embedded throughout our organisation and our community, so that their insights are directly applied to our business. We truly believe in sharing knowledge, so we want to do this without any restrictions.

Read more about Info Support Research here.