Download - Wearable Eye Tracker
Wearable Eye Tracker
Xiaoyong YeFranz Alexander Van Horenbeke
David Abbott
IndexIntroductionBackgroundHardwareSoftware
System DesignAlgorithm
Pupil Localization Ellipse Fitting Calibration Homographic Mapping
Experimental ResultsFuture Work
IntroductionA 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
BackgroundWearable Eye-Tracking information
Who has done previous workWhat they have used
Recent Methods used with eye tracker
ObjectivesHardware
WearableLow-CostLight and
ConfortableMoveable eye-
camera
SoftwareReal-TimeAccurate
HardwareHead-Mounted GearTwo Cameras:
Scene CameraEye Camera
HardwareScene Camera
Captures the scene in front of the user
Fixed to the head
Eye CameraCaptures the eyeWith 5 DOF with
respect to the head
System DesignEye Image Scene Image
Pupil Localization
Ellipse Fitting
Calibration Done?
Mapping
Marker Detection
Calculate Homograp
hy
No
Yes
Ellipse Center
Pupil LocalizationAutomatic Threshold (Modified Otsu’s
Method)
Image Morphology(Dilation, Erosion)
Connected Components Analysis(Find Pupil)
Pupil Center Estimation
Histogram of an Eye Image
Graylevel
Pupil
Background
Threshold
Pupil Localization Threshold
Erosion
Connect Component
s
Pupil Detection
Dilation
Fill holes
Ellipse Fitting1. Updating the pupil Center
2. Need 5 points for Fitting Ellipse model
3. RANSAC to deal with noisy points
Ellipse FittingRANSAC method
Edge Image
Starburst Algorithm
Feature Points
RANSAC
Ellipse Fitting
CalibrationRelationship between Ellipse center to
Scene Image
*=
Scene Position
Homography Pupil Center
Solving for homographies
8 degrees of freedom in 3 x 3 matrix H, so at least n = 8 pairs of 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
1yx
ihgfedcba
wwy'wx'
X’ = Hx
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 ResultsFrame rate 25/second
Accurate Pupil Ellipse
Mapping error is low( 13 pixels in 640*480 image)
DemoLinkhttp://www.youtube.com/watch?v=lBXLpsXB
GOA&context=C25ea4ADOEgsToPDskIo6A6rLXR8eySvaEf82q6h
Future WorkHardware
Lighter camerasScene camera position
Software
Use corneal refletionTry different mapping techniques
Thank you!