- Full text PDF available (10)
A novel method for distinguishing classes of viewers from their ag-gregated eye movements is described. The probabilistic framework accumulates uniformly sampled gaze as Gaussian point spread functions (heatmaps), and measures the distance of unclassified scanpaths to a previously classified set (or sets). A similarity measure is then computed over the… (More)
A GPU implementation is given for real-time visualization of aggregate eye movements (gaze) via heatmaps. Parallelization of the algorithm leads to substantial speedup over its CPU-based implementation and, for the first time, allows real-time rendering of heatmaps atop video. GLSL shader colorization allows the choice of color ramps. Several… (More)
Academia has embarked on a journey where credentials must be shown on a regular basis: publish or perish. Obtaining the ticket to success is beyond quantity, which can be measured; it is related to more intangible terms such as excellence and quality. Research assessment exercises are now here to stay, and they test not only those who wish to jump onboard,… (More)
Research assessment has become one of the most taxing exercises that scholars have to endure on a regular basis. Publications seem to be the currency used nowadays for everything from: getting funds for a PhD, to gaining access to an academic post and indeed promotion. Research assessment results are also of key importance in university departments, as… (More)
Ambient Intelligence is a new paradigm in which environments are sensitive and responsive to the presence of people. This is having an increasing importance in multimedia applications, which frequently rely on sensors to provide useful information to the user. In this context, multimedia applications must adapt and personalize both content and interfaces in… (More)
A machine learning approach used to classify aggregate gaze distributions recorded by an eye tracker and visualized as heatmaps is demonstrated to successfully discriminate between free and task-driven exploration of video clips.
This article aims to present the project " Multimedia Mining System for AVT " (2008MQD000014), financed by " Departament d'Innovació, Universitats i Empresa de la Generalitat de Catalunya " and developed at the Universitat Autònoma de Barcelona from 2008 till 2010. The result of the project is a platform which fosters autonomous learning in the field of… (More)