Xiaoyong Ye Franz Alexander Van Horenbeke David Abbott Index Introduction Background Hardware Software System Design Algorithm Pupil Localization Ellipse Fitting Calibration Homographic Mapping Experimental Results Future Work Introduction A complete system able to track the user’s eye and map the position of their pupil with the area at which they are looking at in the scene in front of them Background Wearable Eye-Tracking information Who has done previous work What they have used Recent Methods used with eye tracker Objectives Hardware Wearable Low-Cost Light and Confortable Moveable eye-camera Software Real-Time Accurate Hardware Head-Mounted Gear Two Cameras: Scene Camera Eye Camera Hardware Scene Camera Eye Camera Captures the scene in Captures the eye front of the user Fixed to the head With 5 DOF with respect to the head System Design Eye Image Pupil Localization Scene Image Calibration Done? No Ellipse Fitting Ellipse Center Marker Detection Calculate Homography Mapping Yes Pupil Localization Automatic Threshold (Modified Otsu’s Method) Image Morphology(Dilation, Erosion) Connected Components Analysis(Find Pupil) Pupil Center Estimation Histogram of an Eye Image Background Pupil Graylevel Threshold Pupil Localization Threshold Erosion Connect Components Pupil Detection Dilation Fill holes Ellipse Fitting 1. Updating the pupil Center 2. Need 5 points for Fitting Ellipse model 3. RANSAC to deal with noisy points Ellipse Fitting Edge Image RANSAC method Starburst Algorithm Feature Points RANSAC Ellipse Fitting Calibration Relationship between Ellipse center to Scene Image * = Scene Position Homography Pupil Center Solving for homographies X’ = Hx wx' a wy' d w g b e h c x f y i 1 8 degrees of freedom in 3 x 3 matrix H, so at least n points are sufficient to determine it Set up a system of linear equations: Ah = 0 where vector of unknowns h = [a,b,c,d,e,f,g,h]T Need at least 8 eqs, but the more the better… Solve for h. solve using least-squares = 8 pairs of calibration method 1. Look at Scene Marker and Press corresponding number on keyboard, 2. Each marker press 2 to 3 times. 3. Randomly select 8 pairs of points to calculate Homography.(Repeatly) 3. Choose the best Homography matrix. Mapping (x1, y1) (x2, y2) Experimental Results Frame rate 25/second Accurate Pupil Ellipse Mapping error is low( 13 pixels in 640*480 image) Demo Link http://www.youtube.com/watch?v=lBXLpsXBGOA&co ntext=C25ea4ADOEgsToPDskIo6A6rLXR8eySvaEf82q 6h Future Work Hardware Lighter cameras Scene camera position Software Use corneal refletion Try different mapping techniques Thank you!