|COOKIES: By using this website you agree that we can place Google Analytics Cookies on your device for performance monitoring.|
Eye Tracking with Consumer Hardware
If you have a question about this talk, please contact Phil Cowans.
Present commercial gaze trackers (i.e. from Tobii and LC Technology) are easy to use, robust and sufficiently accurate for many screen-based applications but their costs exceed the budget of most people. Low cost eye tracking has received an increased attention due to the rapid developments in tracking hardware (video boards, digital camera and CPUs). Eye tracking based on consumer hardware is subject to several unknown factors as various system parameters (i.e. camera parameters and geometry) are unknown. Robust statistical principles to accommodate uncertainties in image data are therefore needed. I will discuss the components (detection, tracking and gaze estimation) used in a low-cost eye tracker. I will in particular describe our contour-based iris tracker. The contour model is based on the statistics of natural images. It turns out that through fairly simple modeling that explicit feature detection can be avoided and thus thresholds become needless. Based on the data from the eye tracker I will then discuss current gaze estimation methods and compare them with gaze estimation methods using Gaussian Processes.
This talk is part of the Inference Group series.
This talk is included in these lists:
Note that ex-directory lists are not shown.
Other listsPolar Humanities Research Group - Scott Polar Research Institute EPRG Energy and Environment (E&E) Series Easter 2012 Interfacial Studies on a Charged Surface in Oil by Contact Angle Measurements
Other talksRecombinant adeno-associated virus for therapeutic gene delivery: 30 million years in the making. De novo Design of Transmembrane Helix-Helix Interactions in Membrane Proteins FORTHCOMING SPEAKERS DATES NOT ARRANGED.. The origin and evolution of Homo sapiens Tasmanian devils, a species threatened by two transmissible cancers The frictional frequency response and model identification