Dr. Tiago H. Falk, Assistant Professor INRS-EMT, University of Quebec, Monteal
Harnessing Neurophysiological Insights to Improve Objective Multimedia Quality-of-Experience (QoE) Perception Models
Today, multimedia content is being delivered over heterogeneous networks and consumed in a wide variety of formats, bit and compression rates, as well as display sizes. In order to remain competitive, service providers have to ensure that end-users are satisfied not only with the quality of the delivered content but also with their perceived Quality-of-Experience (QoE). QoE is defined as the “degree of delight or annoyance of the user of an application, resulting from the fulfillment of his/her expectations in light of the user’s personality and current mental state” and is driven by three ‘influence’ factors (IFs): system, context, and human. System IFs refer to technological factors that can be readily measured from e.g., the transmission network (packet loss rate) or reproduction systems (display size, bitrate). Context IFs describe the user’s environment and can be represented by physical (e.g., location), temporal (time of day), socio-economic (cost), or technical (e.g., smartphone) factors. Lastly, human IFs characterize the user’s emotional/mental state, expectation, and attention, amongst other subjective factors. For much of the last decade, objective QoE models have been developed focusing on system or contextual IFs, and human IFs have been completely overlooked.
In this talk, I will start by showing examples that highlight the importance of subjective factors, such as user engagement, preference, emotions, and expectations into the final QoE perception equation. I will continue by describing the state-of-the-art neurophysiological experimental facility developed at INRS-EMT (Montreal) for QoE assessment. The talk will proceed with an introduction into the neuroimaging modalities used in our studies, namely electroencephalo-graphy (EEG) and near-infrared spectroscopy (NIRS). Lastly, I will describe our ongoing studies that are looking at the effects of preference, emotion, and intelligibility on QoE perception for different audio-visual applications. Ultimately, it is expected that this research will lead to innovative ‘QoE-aware’ schemes that can benefit the entertainment, education, and health fields.
Tiago H. Falk received the B.Sc. degree from the Federal University of Pernambuco, Brazil, in 2002, and the M.Sc. and Ph.D. degrees from Queen's University, Canada, in 2005 and 2008, respectively, all in electrical engineering. From 2009-2010, he was an NSERC Postdoctoral Fellow at the University of Toronto working on neurotechnology development. Since 2010 he has been an Assistant Professor at INRS-EMT (University of Quebec) in Montreal, where he directs the Multimedia Signal Analysis and Enhancement (MuSAE) Lab (http://www.musaelab.ca). The Lab conducts research at the crossroads of biomedical engineering and telecommunications and develops award-winning biologically-inspired signal processing techniques with applications across three themes: multimedia communications, health diagnostics, and human-machine interaction. His research team has published over 130 papers in top-tiered journals, peer-reviewed conferences and international book chapters covering these topics.
Throughout his career, Prof. Falk has received several research excellence awards, including the IEEE Kingston Section Ph.D. Research Excellence Award (2008), the 2009-2010 IBBME Best Paper Award (category: Engineering in a Clinical Setting), the Best Student Paper Awards at the International Conference on Acoustics, Speech, and Signal Processing (2005) and the International Workshop on Acoustic Echo and Noise Control (2008); the Best Paper Award at the Intl. Conf. on Information Science, Signal Processing and their Applications; as well as the Prof. Newton Maia Young Scientist Award (2001). He is a Senior IEEE Member, a Technical Committee Member of the IEEE Speech and Language Technical Committee, as well as a Member of the Editorial Board of the Journal of the Canadian Acoustical Association.
Sponsored by the Centre for Pattern Analysis and Machine Intelligence in association with IEEE Signal Processing, Computational Intelligence and Systems, Man and Cybernetics Society Chapters of KW section.