Eye gaze tracking

1. A computer-implemented method comprising:

directing a user to gaze in a plurality of defined angles;

receiving during a first time period, for each angle in the plurality of defined angles, at least one image of at least one eye of the user gazing in the angle;

creating, for each angle in the plurality of defined angles and based on the at least one image of the at least one eye of the user gazing in the angle, at least one enrollment template comprising eye vascular interest point locations and corresponding local features;

receiving during a second, later time period at least one image of at least one eye of the user;

identifying, for the at least one image of the at least one eye received during the second, later time period, a plurality of stable vascular points associated with a vasculature of the at least one eye;

creating at least one verification template based on the stable vascular points identified in the at least one image received during the second, later time period; and

determining a gaze angle of the at least one eye based on matching the at least one verification template to at least one of the enrollment templates.

2. The method of claim 1 , further comprising:

extracting a plurality of feature descriptors around the identified stable vascular points wherein creating the at least one verification template comprises creating a plurality of templates based on the feature descriptors and identified stable vascular points.

3. The method of claim 1 , further comprising identifying one or more stable peri-ocular anchor points based on one or more of the received images of the at least one eye.

4. The method of claim 3 , further comprising:

tracking one or more of the identified stable vascular points over time; and

determining a gaze angle of the at least one eye based on a displacement of one or more of the identified stable vascular points relative to one or more of the stable peri-ocular anchor points.

5. The method of claim 4 , wherein a particular stable peri-ocular anchor point comprises a corner of the at least one eye.

6. The method of claim 4 , wherein the identified stable vascular points comprises stable vascular points on a left eye and a right eye, and wherein the stable peri-ocular anchor points comprise stable peri-ocular anchor points relative to the left eye and the right eye.

7. The method of claim 1 , further comprising:

enhancing a contrast of a white part of the at least one eye in a particular received image of the at least one eye; and

removing noise from a white part of the at least one eye in a particular received image of the at least one eye.

8. The method of claim 7 , wherein the noise comprises an eyelash.

9. The method of claim 1 , further comprising determining an approximate location of an iris of the at least one eye based on a particular received image of the at least one eye.

10. A system comprising:

one or more computers programmed to perform operations comprising:

directing a user to gaze in a plurality of defined angles;

receiving during a first time period, for each angle in the plurality of defined angles, at least one image of at least one eye of the user gazing in the angle;

creating, for each angle in the plurality of defined angles and based on the at least one image of the at least one eye of the user gazing in the angle, at least one enrollment template comprising eye vascular interest point locations and corresponding local features;

receiving during a second, later time period at least one image of at least one eye of the user;

identifying, for the at least one image of the at least one eye received during the second, later time period, a plurality of stable vascular points associated with a vasculature of the at least one eye;

creating at least one verification template based on the stable vascular points identified in the at least one image received during the second, later time period; and

determining a gaze angle of the at least one eye based on matching the at least one verification template to at least one of the enrollment templates.

11. The system of claim 10 , wherein the operations further comprise:

extracting a plurality of feature descriptors around the identified stable vascular points wherein creating the at least one verification template comprises creating a plurality of templates based on the feature descriptors and identified stable vascular points.

12. The system of claim 10 , wherein the operations further comprise identifying one or more stable peri-ocular anchor points based on one or more of the received images of the at least one eye.

13. The system of claim 12 , wherein the operations further comprise:

tracking one or more of the identified stable vascular points over time; and

determining a gaze angle of the at least one eye based on a displacement of one or more of the identified stable vascular points relative to one or more of the stable peri-ocular anchor points.

14. The system of claim 13 , wherein a particular stable peri-ocular anchor point comprises a corner of the at least one eye.

15. The system of claim 13 , wherein the identified stable vascular points comprises stable vascular points on a left eye and a right eye, and wherein the stable peri-ocular anchor points comprise stable peri-ocular anchor points relative to the left eye and the right eye.

16. The system of claim 10 , wherein the operations further comprise:

enhancing a contrast of a white part of the at least one eye in a particular received image of the at least one eye; and

removing noise from a white part of the at least one eye in a particular received image of the at least one eye.

17. The system of claim 16 , wherein the noise comprises an eyelash.

18. The system of claim 10 , wherein the operations further comprise determining an approximate location of an iris of the at least one eye based on a particular received image of the at least one eye.