Capturing the Singularity of Gaze Behaviour with Hidden Markov Models

Our eyes constantly move around to place our high-resolution fovea on the most relevant visual information. They transmit around 1000 Mbits of data to our brain per second.

Eye movements provide a high-resolution spatio-temporal measure of cognitive and visual processes; they are critical to many scientific fields, from image processing to cognitive and clinical neuroscience. In this seminar I will present data mining techniques based on Hidden Markov Models (HMM) able to capture the highly dynamic and individualistic dimensions of gaze behaviour. HMM can easily be used by any scientist to discover clusters of gaze patterns, classify observers based on their gaze behaviour, or inform the development of models of visual attention tailored for a specific population.