The lecture entitled “Affective Computing: Human Behaviour Analysis Using Computer Vision and Machine Learning”
Social AI agents and social robotic systems have recently drawn increasing attention in many disciplines, such as computer vision, artificial intelligence, human-machine interaction, and smartification of processes, and there has emerged an interdisciplinary undertaking. In his talk, Professor Gholamreza Anbarjafari will discuss computer vision applications in human-robot interaction with focus on affective computing. He will talk about different feature extraction approaches (e.g., visual-based, audio-based, tactile-based), the need for dimensionality reduction and semantic understanding (e.g., face detection, recognition and tracking, audio-visual feature extraction). He will also present his research in human behaviour analysis and how it is being used to improve human-robot interaction by recognizing and analysing compound emotions of human while the person is not cooperative with the acquisition system. He will discuss how advances in machine learning such as support vector machines, Hidden Markov Models, and deep learning techniques, have been used in audio-visual feature extraction and face and body analysis using high spatio-temporal resolution. He will also talk about his ongoing research on developing a computer vision-based solution for understanding how humans are blocking emotions. He will finalise his talk by addressing the issues such as the fairness of AI from computer vision perspective.
Gholamreza Anbarjafari (aka Shahab) is Professor of Computer Vision at the Institute of Technology, University of Tartu. He is founder of the intelligent computer vision (iCV) lab (the largest CV lab in the Baltic states) at the University of Tartu. He was also Deputy Scientific Coordinator of the European Network on Integrating Vision and Language (iV&L Net) ICT COST Action IC1307 and has represented Estonia in several other COST Actions. He is Associate Editor and Guest Lead Editor of several journals, special issues and book projects, including Springer SIVP, MDPI Entropy, MDPI Information, JAD, Springer MVAP, and EURASIP JIVP. He is an IEEE Senior member and the Chair of Signal Processing/Circuits and Systems/Solid-State Circuits Joint Societies Chapter of the IEEE Estonian section. He has held many national and international grants funded by national agencies, EU agencies, and industrial parties. He is expert in computer vision, machine learning, human-robot interaction, graphical models and artificial intelligence. He has supervised 17 MSc students and 7 PhD students and published over 130 scientific works. He has been in the organising and technical committee of many conferences, organised challenges and workshops at various events and been involved in several start-ups and spin-offs.
ALL ARE WELCOME TO ATTEND!
There will be a live webcast of the lecture, which can be viewed on the university’s video portal www.uttv.ee