Determination of gaze calibration parameters

Patent metadata

Field	Value
Publication number	US-11169604-B2
Application number	US-202017099246-A
Country	US
Kind code	B2
Filing date	Nov 16, 2020
Priority date	Nov 15, 2019
Publication date	Nov 9, 2021
Grant date	Nov 9, 2021

Abstract

A method for determining gaze calibration parameters for gaze estimation of a viewer using an eye-tracking system. The method comprises obtaining a set of data points including gaze tracking data of the viewer and position information of at least one target visual; selecting a first subset of the data points and determining gaze calibration parameters using said first subset. A score for the gaze calibration parameters is determined by using the gaze calibration parameters with a second subset of data points, wherein at least one data point of the subset is not included in the first subset. The score is indicative of the capability of the gaze calibration parameters to reflect position information of the at least one target visual based on the gaze tracking data. The score is compared to a candidate score and if it is higher, the calibration parameters are set to the candidate calibration parameters and the score to the candidate score.

First claim

The invention claimed is:

1. A method for determining a set of gaze calibration parameters for gaze estimation of a viewer using an eye-tracking system, wherein the eye-tracking system is associated with a candidate set of gaze calibration parameters, the method comprising: a) obtaining a set of data points at at least one point in time, wherein the data points include gaze tracking data of the viewer and position information of at least one target visual that may attract the gaze of the viewer at said at least one point in time; b) selecting a first subset of said obtained set of data points; c) determining a set of gaze calibration parameters using said first subset of obtained set of data points; d) determining a score for the determined set of gaze calibration parameters by using the set of gaze calibration parameters with a second subset of said obtained set of data points, w herein at least one of the data points in the second subset of data points is not included in the first subset of data points, wherein the score is indicative of the capability of the set of gaze calibration parameters to reflect the position information of the at least one target visual based on the gaze tracking data; e) comparing the score for the set of gaze calibration parameters to a stored candidate score; and f) if the score is higher than the stored candidate score, setting the set of gaze calibration parameters to the candidate set of gaze calibration parameters and the score to the candidate score, wherein the set of gaze calibration parameters minimize the sum of at least one offset value, wherein each offset value indicates an offset between each of the at least one target visual and a gaze estimation comprising at least one estimated gaze point, wherein the gaze estimation is based on said gaze tracking data.

2. The method according to claim 1 , wherein the step of determining a set of gaze calibration parameters comprises optimizing a loss function which determines the at least one offset value, wherein the at least one offset is determined by: a) determining at least one estimated gaze point by inputting the obtained gaze tracking data and the candidate gaze calibration parameters into an eye-tracking algorithm; and b) determining the at least one offset value between the at least one estimated gaze point and the position information of each of the at least one target visual.

3. The method according to claim 1 , wherein the offset value is a Euclidean distance between one of the at least one target visual and the gaze estimation in three dimensions.

4. The method according to claim 1 , wherein the offset value is an angle between an estimated gaze direction, represented by a vector between an origin relative to the viewer and the gaze estimation, and a direction of the target visual, represented by a vector between the origin relative to the viewer and one of the at least one target visual.

5. The method according to claim 1 , wherein the score represents a number of inliers, wherein an inlier is a data point having an offset value below a threshold.

6. The method according to claim 5 , wherein, if the number of inliers is the same for the score and the candidate score, the score is further based on the sum of the offset value of the inliers.

7. A computing device for determining a set of gaze calibration parameters for gaze estimation of a viewer using an eye-tracking system, wherein the eye-tracking system is associated with a candidate set of gaze calibration parameters, the computing device comprising: a processor; and a memory comprising instructions executable by the processor whereby the computing device is operative to: a) obtain a set of data points at at least one point in time, wherein the data points include gaze tracking data of the viewer and position information of at least one target visual that may attract the gaze of the viewer at said at least one point in time; b) select a first subset of said obtained set of data points; c) determine a set of gaze calibration parameters using said first subset of obtained set of data points; d) determine a score for the determined set of gaze calibration parameters by using the set of gaze calibration parameters with a second subset of said obtained set of data points, wherein at least one of the data points in the second subset of data points is not included in the first subset of data points, wherein the score is indicative of the capability of the set of gaze calibration parameters to reflect the position information of the at least one target visual based on the gaze tracking data; e) compare the score for the set of gaze calibration parameters to a stored candidate score; and f) if the score is higher than the stored candidate score, set the set of gaze calibration parameters to the candidate set of gaze calibration parameters and the score to the candidate score, wherein the set of gaze calibration parameters minimize the sum of at least one offset value, wherein each offset value indicates an offset between each of the at least one target visual and a gaze estimation comprising at least one estimated gaze point, wherein the gaze estimation is based on said gaze tracking data.

8. The computing device according to claim 7 , wherein the memory comprises instructions executable by the processor whereby the computing device is operative to determine a set of gaze calibration parameters by optimizing a loss function, which determines the at least one offset value, wherein the computing device is operative to determine the at least one offset by: a. determine at least one estimated gaze point by inputting the obtained gaze tracking data and the candidate gaze calibration parameters into an eye-tracking algorithm; and b. determine the at least one offset value between the at least one estimated gaze point and the position information of each of the at least one target visual.

9. The computing device according to claim 7 , wherein the offset value is a Euclidean distance between one of the at least one target visual and the gaze estimation in three dimensions.

10. The computing device according to claim 7 , wherein the offset value is an angle between an estimated gaze direction, represented by a vector between an origin relative to the viewer and the gaze estimation, and a direction of the target visual, represented by a vector between the origin relative to the viewer and one of the at least one target visual.

11. The computing device according to claim 8 , wherein the score represents a number of inliers, wherein an inlier is a data point having an offset value below a threshold.

12. The computing device according to claim 11 , wherein if the number of inliers is the same for the score and the candidate score, the first score is further based on the sum of the offset value of the inliers.

13. A non-transitory computer readable storage medium comprising a computer program, the computer program comprising instructions which when executed on at least one processor, cause the at least one processor to carry out the method comprising: a) obtaining a set of data points at at least one point in time, wherein the data points include gaze tracking data of the viewer and position information of at least one target visual that may attract the gaze of the viewer at said at least one point in time; b) selecting a first subset of said obtained set of data points; c) determining a set of gaze calibration parameters using said first subset of obtained set of data points; d) determining a score for the determined set of gaze calibration parameters by using the set of gaze calibration parameters with a second subset of said obtained set of data points, wherein

Assignees

• Tobii Ab

Inventors

• Barkman Patrik
• Molin David

Classifications

• G06V40/193

  Preprocessing; Feature extraction · CPC title

• G06V40/19

  Sensors therefor · CPC title

• G06F3/013 Primary

  Eye tracking input arrangements (G06F3/015 takes precedence) · CPC title

• G06K9/00604

  Physics · mapped topic