Multi-modal Emotion Recognition using facial expression and other physiological condition

Developing a system (or machine learning model) that can try to predict or confirm the emotional mind state of a subject given additional information like pulse rate blood pressure etc.
Master

Emotion Recognition has been largely explored for its application like in video analytics. Traditional emotion recognition system focuses on visual facial expressions for recognizing emotion of the concerned subject. However, this master thesis will aim to map the relationship between emotion of a person along with the subject’s then physiological condition like pulse rate, blood pressure etc. For emotion recognition we will use an object detection based approach.

Goal

A system (or machine learning model) that can try to predict or confirm the emotional mind state of a subject given additional information like pulse rate blood pressure etc.

Learning outcome

Image processing, machine learning, real-world implementation and experiments

Qualifications

  • Python programming, OpenCV, Keras , Tensorflow
  • Knowledge about deep learning and object detection algorithm’s is an advantage

Supervisors

  • Michael Riegler
  • Sukalpa Chanda, (Østfold University College, Norway)
  • Umapada Pal (Indian Statistical Institute, India)

Collaboration partners

  • Indian Statistical Institute
  • Østfold University College, Norway