Angular contact geometry

1. A method, comprising:

recognizing touch input sensor data as a touch input on a touch-screen display;

generating a sensor map from the touch input sensor data, the sensor map representing the touch input;

modeling the sensor map as a Gaussian distribution;

determining an ellipse that approximately encompasses the sensor map from the Gaussian distribution, the ellipse representing a contact shape of the touch input; and

validating the ellipse based on an inaccuracy ratio (IAR) of the ellipse, the IAR of the ellipse determined from one or more regions that are within the bounds of the ellipse, but are not part of the sensor map.

2. A method as recited in claim 1 , wherein:

the contact shape of the touch input is irregular in shape; and

the ellipse is of a size and rotation angle that approximately encompasses the sensor map.

3. A method as recited in claim 1 , wherein the sensor map is generated as a two-dimensional array of elements that correlate to sensed contact from the touch input on the touch-screen display.

4. A method as recited in claim 3 , wherein a stronger sensor signal of the touch input sensor data indicates more contact with an element in the sensor map.

5. A method as recited in claim 1 , further comprising:

determining a minor axis and a major axis of the ellipse from a covariance matrix; and

scaling axis vectors of the minor axis and the major axis to correlate a geometry of the ellipse with the contact shape of the touch input.

6. A method as recited in claim 1 , wherein a best-fit ellipse has an optimal IAR of the smallest value and encompasses most elements of the sensor map with the least non-overlapping region area.

7. A method as recited in claim 6 , further comprising determining a normalized IAR based on the determined IAR and the optimal IAR.

8. A computing device, comprising:

a touch-screen display;

a touch input module configured to recognize touch input sensor data as a touch input on the touch-screen display;

at least a memory and a processor to implement a contact geometry service configured to:

generate a sensor map from the touch input sensor data, the sensor map representing the touch input and modeled as a Gaussian distribution;

determine an ellipse that approximately encompasses the sensor map from the Gaussian distribution, the ellipse representing a contact shape of the touch input; and

validate the ellipse based on an inaccuracy ratio (IAR) of the ellipse, the IAR of the ellipse being determinable from one or more regions that are within the bounds of the ellipse, but are not part of the sensor map.

9. A computing device as recited in claim 8 , wherein:

the contact shape of the touch input is irregular in shape; and

the ellipse is of a size and rotation angle that approximately encompasses the sensor map.

10. A computing device as recited in claim 8 , wherein the contact geometry service is further configured to generate the sensor map as a two-dimensional array of elements that correlate to sensed contact from the touch input on the touch-screen display.

11. A computing device as recited in claim 10 , wherein a stronger sensor signal of the touch input sensor data indicates more contact with an element in the sensor map.

12. A computing device as recited in claim 8 , wherein the contact geometry service is further configured to:

determine a minor axis and a major axis of the ellipse from a covariance matrix; and

scale axis vectors of the minor axis and the major axis to correlate a geometry of the ellipse with the contact shape of the touch input.

13. A computing device as recited in claim 12 , wherein the contact geometry service is configured to said scale the axis vectors using a global scaling factor and scaling factors determined from the covariance matrix.

14. A method, comprising:

generating a sensor map from touch input sensor data that is received as a touch input, the sensor map generated as a two-dimensional array of elements that correlate to sensed contact from the touch input;

modeling the sensor map as a Gaussian distribution with variables computed based on a weighted average;

determining an ellipse from the Gaussian distribution, the ellipse representing a contact shape of the touch input and approximately encompassing the elements of the sensor map; and

validating the ellipse based on an inaccuracy ratio (IAR) of the ellipse, the IAR of the ellipse determined from one or more regions that are within the bounds of the ellipse, but are not part of the sensor map.

15. A method as recited in claim 14 , further comprising generating HID reports (human interface device reports) that include a time, position data, and angular contact geometry data that correlates to the touch input.

16. A method as recited in claim 14 , wherein a best-fit ellipse has an optimal IAR of the smallest value and encompasses most of the elements of the sensor map with the least non-overlapping region area.

17. A method as recited in claim 16 , further comprising determining a normalized IAR based on the determined IAR and the optimal IAR, and wherein said determining the optimal IAR is performed off-line.

18. A method as recited in claim 14 , further comprising representing the ellipse as a rectangle that bounds the ellipse, the rectangle defined by a height, a width, and a rotation angle.

19. A method as recited in claim 18 , further comprising:

determining a minor axis and a major axis of the rectangle that represents the ellipse; and

scaling axis vectors of the minor axis and the major axis to correlate a geometry of the ellipse with the contact shape of the touch input.

20. A method as recited in claim 14 , wherein:

the contact shape of the touch input is irregular in shape; and

the ellipse is of a size and rotation angle that approximately encompasses the sensor map.