Evaluation method and system for tracking angle range of eye tracking device

Abstract

The present invention discloses a method and system for evaluating the tracking angle range of an eye movement tracking device. An artificial eye is provided inside the eye movement tracking device. The method includes: controlling the artificial eye to continuously rotate with the center point as the starting measurement point along a preset measurement direction at a preset angle gradient value, collecting raw eye movement data through the eye movement tracking device, and stopping the rotation of the artificial eye and data collection when the data loss rate reaches a preset value; calculating the data loss rate of each tracking angle in the raw eye movement data, and determining the partition angles of the medium and low recognition areas of eye movement tracking in the corresponding measurement direction of the eye movement tracking device according to the data loss rate; calculating the accuracy of each tracking angle in the raw eye movement data, and determining the partition angles of the medium and high recognition areas of eye movement tracking in the corresponding measurement direction of the eye movement tracking device according to the accuracy. The present invention can calibrate the tracking range of the eye movement tracking system and is applicable to head-mounted eye movement tracking devices.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual reality, and particularly to a method and system for evaluating the tracking angle range of an eye movement tracking device. Background Art

[0002] An eye movement tracking device is one of the commonly used devices in virtual reality technology. In the current technology, there is no systematic method for evaluating the eye movement tracking angle range. There are only methods for measuring the accuracy at specific tracking angles, including experimental measurements with real people and measurements with artificial eyes.

[0003] Among them, the existing method for measuring with an artificial eye is only applicable to a telemetric eye tracker and is not suitable for measuring the accuracy of a head-mounted eye movement tracking device. In addition, these two types of methods only measure the accuracy at different tracking angles and do not further calibrate the tracking range of the eye movement tracking system, restricting the further application and development of eye movement interaction. Summary of the Invention

[0004] Therefore, an embodiment of the present invention provides a method for evaluating the tracking angle range of an eye movement tracking device to calibrate the tracking range of the eye movement tracking system and is applicable to a head-mounted eye movement tracking device.

[0005] According to an embodiment of the present invention, in the method for evaluating the tracking angle range of an eye movement tracking device, an artificial eye is provided inside the eye movement tracking device, and the method includes:

[0006] Controlling the artificial eye to continuously rotate with the center point as the starting measurement point along a preset measurement direction at a preset angle gradient value, collecting raw eye movement data through the eye movement tracking device, and stopping rotating the artificial eye and collecting data when the data loss rate reaches a preset value;

[0007] Calculating the data loss rate of each tracking angle in the raw eye movement data, and determining the partition angles of the medium and low recognition areas of eye movement tracking of the eye movement tracking device in the corresponding measurement direction according to the data loss rate;

[0008] Calculating the accuracy of each tracking angle in the raw eye movement data, and determining the partition angles of the medium and high recognition areas of eye movement tracking of the eye movement tracking device in the corresponding measurement direction according to the accuracy.

[0009] The method for evaluating the tracking angle range of an eye movement tracking device according to an embodiment of the present invention first controls an artificial eye to continuously rotate with the center point as the starting measurement point along a preset measurement direction at a preset angle gradient value, so that the original eye movement data can be collected by the eye movement tracking device. Then, the data loss rate is used to determine the partition angles of the medium and low recognition areas in the eye movement tracking of the eye movement tracking device in the corresponding measurement direction, and the accuracy is used to determine the partition angles of the medium and high recognition areas in the eye movement tracking of the eye movement tracking device in the corresponding measurement direction, realizing the calibration of the tracking range and being applicable to a head-mounted eye movement tracking device. The present invention can provide support for subsequent eye movement interaction applications.

[0010] In addition, the method for evaluating the tracking angle range of an eye movement tracking device according to the above embodiment of the present invention may further have the following additional technical features:

[0011] Further, the step of calculating the data loss rate of each tracking angle in the original eye movement data and determining the partition angles of the medium and low recognition areas in the eye movement tracking of the eye movement tracking device in the corresponding measurement direction according to the data loss rate specifically includes:

[0012] Traverse each tracking angle in the original eye movement data in descending order of the absolute value of the tracking angle. When the data loss rate of the first tracking angle is less than or equal to the preset loss rate, and the data loss rate of the next tracking angle greater than the first tracking angle is greater than the preset loss rate, then determine the first tracking angle as the partition angle of the medium and low recognition areas in the eye movement tracking of the eye movement tracking device in the measured direction.

[0013] Further, in the step of calculating the data loss rate of each tracking angle in the original eye movement data, for the first target tracking angle, the data loss rate of the first target tracking angle is calculated by the following formula:

[0014] Loss = N 无效 / (f * t);

[0015] where Loss represents the data loss rate of the first target tracking angle, N 无效 represents the number of invalid samples corresponding to the first target tracking angle, f represents the sampling frequency of the eye movement tracking device, and t represents the total sampling duration corresponding to the first target tracking angle.

[0016] Further, the step of calculating the accuracy of each tracking angle in the original eye movement data and determining the partition angles of the medium and high recognition areas in the eye movement tracking of the eye movement tracking device in the corresponding measurement direction according to the accuracy specifically includes:

[0017] Calculate the accuracy of each tracking angle in the original eye movement data, and perform a normal transformation on the accuracy to obtain the transformed accuracy of each tracking angle;

[0018] Calculate the critical value based on the transformed accuracy of all tracking angles;

[0019] Traverse each tracking angle in the original eye movement data in descending order of the absolute value of the tracking angle. When the transformed accuracy of the second tracking angle is less than or equal to the critical value, and the transformed accuracies of all tracking angles greater than the second tracking angle are greater than the critical value, then determine that the second tracking angle is the partitioning angle of the medium and high recognition areas of the eye movement tracking in the measured direction by the eye movement tracking device, and the second tracking angle is less than the first tracking angle.

[0020] Further, the steps of calculating the accuracy of each tracking angle in the original eye movement data and performing a normal transformation on the accuracy to obtain the transformed accuracy of each tracking angle specifically include:

[0021] For the second target tracking angle, calculate the angle formed by the set visual stimulus point and the fixation point detected by the eye movement tracking device with the pupil center point, and calculate the accuracy of the second target tracking angle based on the angle;

[0022] Use the Boxcox method to perform a normal transformation on the accuracy of the second target tracking angle to obtain the transformed accuracy of the second target tracking angle.

[0023] Further, in the step of calculating the angle formed by the set visual stimulus point and the fixation point detected by the eye movement tracking device with the pupil center point for the second target tracking angle and calculating the accuracy of the second target tracking angle based on the angle, the following formula is used to calculate the accuracy of the second target tracking angle:

[0024]

[0025]

[0026] where A represents the accuracy of the second target tracking angle, n represents the total number of samples, i represents the i-th moment, α i represents the angle, abs represents the absolute value operation, represents the line of sight vector formed by the visual stimulus point on the display screen of the eye movement tracking device and the pupil center of the artificial eye at the i-th moment, is the line of sight vector detected by the eye movement tracking device at the i-th moment.

[0027] Further, in the step of calculating the critical value according to the converted accuracies of all tracking angles, the critical value is calculated using the following formula:

[0028] L limit = mean - k * SD

[0029] where L limit represents the critical value, mean represents the mean of the converted accuracies of all tracking angles, SD represents the standard deviation of the converted accuracies of all tracking angles, and k represents the correction coefficient.

[0030] Further, when the data loss rate reaches a preset value, the steps of stopping the rotation of the artificial eye and collecting data specifically include:

[0031] When the artificial eye rotates to the target tracking angle and the next two consecutive angles, such that the eye movement tracking device fails to collect eye movement data at these three consecutive angles, i.e., the data loss rate reaches 100%, at this time, stop the rotation of the artificial eye and collecting data. The angle before the target tracking angle is the maximum angle that the eye movement tracking device can track.

[0032] Another embodiment of the present invention provides a tracking angle range evaluation system for an eye movement tracking device to calibrate the tracking range of an eye movement tracking system and is applicable to a head-mounted eye movement tracking device.

[0033] According to the tracking angle range evaluation system of an eye movement tracking device according to an embodiment of the present invention, the eye movement tracking device is provided with an artificial eye, and the system includes:

[0034] A control acquisition module for controlling the artificial eye to continuously rotate around the center point as the starting measurement point along a preset measurement direction with a preset angle gradient value, collecting raw eye movement data through the eye movement tracking device, and stopping the rotation of the artificial eye and collecting data when the data loss rate reaches a preset value;

[0035] A first calculation module for calculating the data loss rate of each tracking angle in the raw eye movement data and determining the partition angles of the medium and low recognition areas of eye movement tracking in the corresponding measurement direction of the eye movement tracking device according to the data loss rate;

[0036] A second calculation module for calculating the accuracy of each tracking angle in the raw eye movement data and determining the partition angles of the medium and high recognition areas of eye movement tracking in the corresponding measurement direction of the eye movement tracking device according to the accuracy.

[0037] The tracking angle range evaluation system of the eye movement tracking device according to an embodiment of the present invention first controls the artificial eye to continuously rotate with the center point as the starting measurement point along the preset measurement direction at a preset angle gradient value, and can collect the original eye movement data through the eye movement tracking device. Then, the data loss rate is used to determine the partition angles of the medium and low recognition areas of the eye movement tracking in the corresponding measurement direction of the eye movement tracking device, and the partition angles of the medium and high recognition areas of the eye movement tracking in the corresponding measurement direction of the eye movement tracking device are determined according to the accuracy, realizing the calibration of the tracking range, and being applicable to the head-mounted eye movement tracking device. The present invention can provide support for subsequent eye movement interaction applications.

[0038] In addition, the tracking angle range evaluation system of the eye movement tracking device according to the above embodiment of the present invention may further have the following additional technical features:

[0039] Further, the first calculation module is specifically configured to:

[0040] Traverse each tracking angle in the original eye movement data in descending order of the absolute value of the tracking angle. When the data loss rate of the first tracking angle is less than or equal to the preset loss rate, and the data loss rate of the next tracking angle greater than the first tracking angle is greater than the preset loss rate, then it is determined that the first tracking angle is the partition angle of the medium and low recognition areas of the eye movement tracking of the eye movement tracking device in the measured direction.

[0041] Further, the first calculation module is used to calculate the data loss rate of the first target tracking angle by using the following formula:

[0042] Loss = N 无效 / (f * t);

[0043] where Loss represents the data loss rate of the first target tracking angle, N 无效 represents the number of invalid samples corresponding to the first target tracking angle, f represents the sampling frequency of the eye movement tracking device, and t represents the total sampling duration corresponding to the first target tracking angle.

[0044] Further, the second calculation module is specifically configured to:

[0045] Calculate the accuracy of each tracking angle in the original eye movement data, and perform a normal transformation process on the accuracy to obtain the transformed accuracy of each tracking angle;

[0046] Calculate the critical value according to the transformed accuracy of all tracking angles;

[0047] Traverse each tracking angle in the original eye movement data in descending order of the absolute value of the tracking angle. When the converted accuracy of the second tracking angle is less than or equal to the critical value, and the converted accuracies of all tracking angles greater than the second tracking angle are greater than the critical value, then determine that the second tracking angle is the partition angle of the medium and high recognition areas in the eye movement tracking of the eye movement tracking device in the measured direction, and the second tracking angle is less than the first tracking angle.

[0048] Further, the second calculation module is specifically configured to:

[0049] For the second target tracking angle, calculate the included angle formed by the set visual stimulus point, the fixation point detected by the eye movement tracking device, and the pupil center point, and calculate the accuracy of the second target tracking angle according to the included angle;

[0050] Use the Boxcox method to perform normal transformation processing on the accuracy of the second target tracking angle to obtain the converted accuracy of the second target tracking angle.

[0051] Further, the second calculation module is specifically configured to calculate the accuracy of the second target tracking angle using the following formula:

[0052]

[0053]

[0054] where A represents the accuracy of the second target tracking angle, n represents the total number of samples, i represents the i-th moment, and α i represents the included angle, abs represents the absolute value operation, represents the line-of-sight vector formed by the visual stimulus point on the display screen of the eye movement tracking device and the pupil center of the artificial eye at the i-th moment, is the line-of-sight vector detected by the eye movement tracking device at the i-th moment.

[0055] Further, the second calculation module is specifically configured to calculate the critical value using the following formula:

[0056] L limit = mean - k * SD

[0057] where L limit represents the critical value, mean represents the mean of the converted accuracies of all tracking angles, SD represents the standard deviation of the converted accuracies of all tracking angles, and k represents the correction coefficient.

[0058] Further, the control acquisition module is specifically configured to:

[0059] When the artificial eye rotates to the target tracking angle and the next two consecutive angles, and the eye movement tracking device fails to collect eye movement data at these three consecutive angles, that is, the data loss rate reaches 100%, stop rotating the artificial eye and collecting data at this time. The angle immediately preceding the target tracking angle is the maximum angle that the eye movement tracking device can track. Brief Description of the Drawings

[0060] The above and / or additional aspects and advantages of the embodiments of the present invention will become apparent and be readily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0061] Figure 1 is a flowchart of a method for evaluating the tracking angle range of an eye movement tracking device according to an embodiment of the present invention;

[0062] Figure 2 is Figure 1 a detailed flowchart of step S103 in

[0063] Figure 3 a schematic diagram of the partition of the eye movement tracking range;

[0064] Figure 4 is a schematic structural diagram of a system for evaluating the tracking angle range of an eye movement tracking device according to an embodiment of the present invention. Detailed Embodiments

[0065] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0066] Please refer to Figure 1 , a method for evaluating the tracking angle range of an eye movement tracking device proposed in an embodiment of the present invention. An artificial eye is provided inside the eye movement tracking device. The method includes steps S101 to S103:

[0067] S101, controlling the artificial eye to continuously rotate with the center point as the starting measurement point along a preset measurement direction at a preset angle gradient value, collecting raw eye movement data through the eye movement tracking device, and stopping rotating the artificial eye and collecting data when the data loss rate reaches a preset value.

[0068] The eye movement tracking device in this embodiment is a head-mounted eye movement tracking device, which is internally provided with an artificial eye. First, taking the geometric center of any left or right lens of the eye movement tracking device as 0°, the center of the artificial eye is made to be on the principal optical axis of the lens, and the distance between the center of the artificial eye and the lens is made equal to the exit pupil distance specified by the manufacturer of the eye movement tracking device. Then, the artificial eye is continuously rotated along a preset measurement direction (such as clockwise or counterclockwise) with a preset angular gradient value. Preferably, the preset angular gradient value is not greater than 1°.

[0069] When the artificial eye rotates to the target tracking angle ω1 and the next two consecutive angles ω2 and ω3, such that the eye movement tracking device fails to collect eye movement data at these three consecutive angles (i.e., ω1, ω2, ω3), that is, the data loss rate reaches 100%, at this time, stop rotating the artificial eye and collecting data, and the angle Ang immediately preceding the target tracking angle max is the maximum angle that the eye movement tracking device can track.

[0070] S102, calculate the data loss rate of each tracking angle in the raw eye movement data, and determine the partition angles of the medium and low recognition regions of eye movement tracking of the eye movement tracking device in the corresponding measurement direction according to the data loss rate.

[0071] Among them, after obtaining the raw eye movement data, it is necessary to perform data processing on the raw eye movement data.

[0072] When the eye movement tracking device fails to successfully collect eye movement data, the pupil size item in the exported data is lost, which is manifested as an abnormal value, and the data sample at this moment is an invalid sample. It is necessary to calculate the ratio of the total amount of invalid samples at each tracking angle to the total amount of data sampling at the corresponding tracking angle to obtain the data loss rate at the corresponding tracking angle.

[0073] Specifically, in the step of calculating the data loss rate of each tracking angle in the raw eye movement data, for the first target tracking angle, the following formula is used to calculate the data loss rate of the first target tracking angle:

[0074] Loss = N 无效 / (f * t);

[0075] where Loss represents the data loss rate of the first target tracking angle, N 无效 represents the amount of invalid samples corresponding to the first target tracking angle, f represents the sampling frequency of the eye movement tracking device, and t represents the total sampling duration corresponding to the first target tracking angle.

[0076] Through the above calculation formula, the data loss rate of each tracking angle in the raw eye movement data can be obtained.

[0077] Then, traverse each tracking angle in the original eye movement data in descending order of the absolute value of the tracking angle. When the data loss rate of the first tracking angle is less than or equal to the preset loss rate, and the data loss rate of the next tracking angle greater than the first tracking angle is greater than the preset loss rate (specifically, here it means that the data loss rate of the next tracking angle whose absolute value is greater than the absolute value of the first tracking angle is greater than the preset loss rate. For example, if the first tracking angle is 45°, the next tracking angle whose absolute value is greater than the absolute value of the first tracking angle is 46°; another example, if the first tracking angle is -45°, the next tracking angle whose absolute value is greater than the absolute value of the first tracking angle is -46°), then determine the first tracking angle as the partition angle of the medium and low recognition areas in the eye movement tracking of the eye movement tracking device in the measured direction. Preferably, the preset loss rate is 20%. In specific implementation, the data loss rate starts from 0° and gradually increases as the absolute value of the tracking angle increases. Therefore, traverse each tracking angle in the original eye movement data in descending order of the absolute value of the tracking angle. When the data loss rate of the first tracking angle is less than or equal to 20%, and the data loss rate of the next tracking angle greater than the first tracking angle is greater than 20%, then determine the first tracking angle as the partition angle of the medium and low recognition areas in the eye movement tracking of the eye movement tracking device in the measured direction.

[0078] S103. Calculate the accuracy of each tracking angle in the original eye movement data, and determine the partition angle of the medium and high recognition areas in the eye movement tracking of the eye movement tracking device in the corresponding measured direction according to the accuracy.

[0079] In specific implementation, the initial fluctuating part of the sampling data at each tracking angle can be removed first to ensure that the remaining data for analysis is the data when the artificial eye has stabilized, and then the accuracy at each tracking angle is calculated.

[0080] Please refer to Figure 2 , step S103 specifically includes S1031 to S1033:

[0081] S1031. Calculate the accuracy of each tracking angle in the original eye movement data, and perform a normal transformation process on the accuracy to obtain the transformed accuracy of each tracking angle.

[0082] Step S1031 specifically includes:

[0083] For the second target tracking angle, calculate the included angle formed by the set visual stimulus point and the fixation point detected by the eye movement tracking device and the center point of the pupil, and calculate the accuracy of the second target tracking angle according to the included angle;

[0084] Among them, the following formula is specifically used to calculate the accuracy of the second target tracking angle:

[0085]

[0086]

[0087] Among them, A represents the accuracy of the second target tracking angle, n represents the total number of samples, i represents the i-th moment, and α i represents the included angle, abs represents the absolute value operation, represents the line-of-sight vector formed by the visual stimulus point on the display screen of the eye tracking device at the i-th moment and the pupil center of the artificial eye, is the line-of-sight vector detected by the eye tracking device at the i-th moment.

[0088] Thus, the accuracy of all tracking angles can be calculated.

[0089] It should be noted that in this embodiment, the accuracy is used to describe the deviation of the eye tracking system in capturing eye movement signals. For example, a coordinate point (point A) is set on the display screen, and the coordinate point (point B) of the eye fixation collected when the human eye or the artificial eye is aligned with this coordinate point is used to calculate the accuracy through the coordinates of these two points. The unit of accuracy is degree. The larger the value of the accuracy, the farther apart points A and B are, that is, the less accurate; on the contrary, the smaller the value of the accuracy, the closer points A and B are, that is, the more accurate.

[0090] Then, the Boxcox method is used to perform a normal transformation on the accuracy of the second target tracking angle to obtain the transformed accuracy of the second target tracking angle. Thus, the transformed accuracy of all tracking angles can be calculated.

[0091] S1032. Calculate the critical value according to the transformed accuracy of all tracking angles;

[0092] Among them, the following formula is specifically used to calculate the critical value:

[0093] L limit = mean - k * SD

[0094] Among them, L limit represents the critical value, mean represents the mean of the transformed accuracy of all tracking angles, SD represents the standard deviation of the transformed accuracy of all tracking angles, and k represents the correction coefficient.

[0095] Preferably, k is 0.5, that is, the value 0.5 standard deviations below the mean is selected as the critical value L corresponding to the medium and high recognition discrimination zone angles limit .

[0096] S1033. Traverse each tracking angle in the original eye movement data in descending order of the absolute value of the tracking angle. When the converted accuracy of the second tracking angle is less than or equal to the critical value, and the converted accuracies of all tracking angles greater than (also referring to greater than in absolute value) the second tracking angle are greater than the critical value, then determine the second tracking angle as the partition angle of the medium and high recognition areas of eye movement tracking in the measured direction by this eye movement tracking device, and the second tracking angle is less than the first tracking angle.

[0097] The following uses a specific example to illustrate the above method. The eye movement tracking angle range of an eye movement tracking device (HPOmnicept Reverb G2 virtual reality headset device) is measured using an artificial eye, and the eye movement tracking component is Tobii XR. The tracking angle range of the eye movement tracking device in the horizontal direction is measured and the recognition area is divided.

[0098] First, with the center of the Fresnel lens on one side of the eye movement tracking device as 0°, the tracking angle is preset to range from -55° (left) to 55° (right), and it is rotated with a preset angle gradient value of 1°. The eye movement data of a total of 111 fixation point angles including 0° is measured, and the sampling duration for each test point is 10s.

[0099] Next, calculate the data loss rate:

[0100] When the eye movement tracking system fails to successfully collect eye movement data, the pupil item in its exported data is -1. Calculate the percentage of the sample size that fails to successfully collect data at each tracking angle in the total 10s sampling data volume, thereby obtaining the data loss rate.

[0101] Then, clear the invalid data and calculate the accuracy:

[0102] Considering the possible vibration error caused by the rotation of the motor, remove the data in the first 2s of the 10s; then remove the eye movement data that fails to be successfully collected according to the pupil data, and calculate the accuracy at each tracking angle after clearing.

[0103] Finally, perform eye movement tracking recognition area division based on the data loss rate and accuracy:

[0104] The data loss rates at each tracking angle on the left and right sides in the horizontal direction, as well as the accuracy obtained after data screening, are shown in Table 1 and Table 2 respectively. The blank value represents no remaining data after screening, the data loss rate is 100%, and the mean and standard deviation cannot be calculated.

[0105] Table 1 Accuracy and data loss rate at different tracking angles on the left

[0106]

[0107]

[0108] Accuracy and data loss rate at different tracking angles on the right side of Table 2

[0109]

[0110]

[0111] Medium and low recognition partition angles:

[0112] According to the data in Table 1 and Table 2, taking the preset loss rate of 20% as the division criterion, the partition angles of the medium and low recognition areas on the right side and the left side can be obtained as 49° on the right side and 47° on the left side respectively.

[0113] Medium and high recognition partition angles:

[0114] First step, perform histogram statistics on the accuracy on both sides of the center point respectively, and its distribution shows a skewed distribution.

[0115] Second step, first delete the null value items of the accuracy, and then use Boxcox to perform normal transformation on the accuracy data.

[0116] Third step, for the right side of the center point, calculate the mean and standard deviation SD after transformation. Among them, mean = 0.960, standard deviation SD = 0.647. Calculate L limit = mean - 0.5SD = 0.637. In the transformed Table 3, when the angle is greater than 23°, the corresponding transformed accuracy is greater than L limit , so 23° is the medium and high recognition partition angle on the right side.

[0117] Fourth step, for the left side of the center point, similar to the third step, mean = 1.115, standard deviation SD = 0.712. L limit = mean - 0.5SD = 0.759. In the transformed Table 4, when the angle is greater than 22°, the corresponding transformed accuracy is greater than L limit , so 22° is the medium and high recognition partition angle on the left side.

[0118] Accuracy and transformed accuracy at different tracking angles on the right side of Table 3

[0119]

[0120]

[0121] Accuracy and transformed accuracy at different tracking angles on the left side of Table 4

[0122]

[0123]

[0124] After partitioning the eye movement tracking range, the schematic diagram of the result is as Figure 3 shown, Figure 3 in which the vertical coordinate on the left side is the accuracy, and the vertical coordinate on the right side is the data loss rate. Curve a is the curve of the accuracy changing with the tracking angle (corresponding to the scale of the left vertical coordinate), and curve b is the curve of the data loss rate changing with the tracking angle (corresponding to the scale of the right vertical coordinate). The abscissa represents the tracking angle from left to right. When the abscissa is around 0°, the tracking effect is the best, and both the accuracy and the data loss rate are the smallest.

[0125] Therefore, based on the above analysis, the conclusion is drawn that the maximum range of Tobii XR eye movement tracking in the HP Omnicept Reverb G2 virtual reality headset device is 52° on the left side and 54° on the right side; its low recognition area is from 47° to 52° on the left side and from 49° to 54° on the right side; the medium recognition area is from 22° to 46° on the left side and from 23° to 48° on the right side; the high recognition area is from 0° to 21° on the left side and from 0° to 22° on the right side.

[0126] The function of dividing the recognition area can be applied to the design of eye movement interaction, for example, in a game that uses eye movement operations. When high precision is required, when designing game elements, it needs to be designed in the high recognition area; if the precision requirement is not so high, it can be designed in the medium recognition area, and so on.

[0127] In summary, according to the method for evaluating the tracking angle range of the eye movement tracking device provided in this embodiment, first, the artificial eye is controlled to continuously rotate along the preset measurement direction with the center point as the starting measurement point and at the preset angle gradient value, and the original eye movement data can be collected by the eye movement tracking device. Then, the data loss rate is used to determine the partition angles of the medium and low recognition areas of the eye movement tracking in the corresponding measurement direction of the eye movement tracking device, and the accuracy is used to determine the partition angles of the medium and high recognition areas of the eye movement tracking in the corresponding measurement direction of the eye movement tracking device, realizing the calibration of the tracking range, and it is applicable to the head-mounted eye movement tracking device. The present invention can provide support for subsequent eye movement interaction applications.

[0128] Please refer to Figure 4 , based on the same inventive concept, an embodiment of the present invention provides a system for evaluating the tracking angle range of an eye movement tracking device. The eye movement tracking device is internally provided with an artificial eye, and the system includes:

[0129] The control acquisition module 10 is used to control the artificial eye to continuously rotate with the center point as the starting measurement point and along the preset measurement direction at a preset angular gradient value, collect the original eye movement data through the eye movement tracking device, and stop rotating the artificial eye and collecting data when the data loss rate reaches the preset value;

[0130] The first calculation module 20 is used to calculate the data loss rate of each tracking angle in the original eye movement data, and determine the partition angles of the medium and low recognition areas of the eye movement tracking in the corresponding measurement direction according to the data loss rate;

[0131] The second calculation module 30 is used to calculate the accuracy of each tracking angle in the original eye movement data, and determine the partition angles of the medium and high recognition areas of the eye movement tracking in the corresponding measurement direction according to the accuracy.

[0132] In this embodiment, the first calculation module 20 is specifically used for:

[0133] Traverse each tracking angle in the original eye movement data in descending order of the absolute value of the tracking angle. When the data loss rate of the first tracking angle is less than or equal to the preset loss rate, and the data loss rate of the next tracking angle greater than the first tracking angle is greater than the preset loss rate, then determine the first tracking angle as the partition angle of the medium and low recognition areas of the eye movement tracking in the measured direction.

[0134] In this embodiment, the first calculation module 20 is used to calculate the data loss rate of the first target tracking angle by using the following formula:

[0135] Loss = N 无效 / (f * t);

[0136] where Loss represents the data loss rate of the first target tracking angle, N 无效 represents the number of invalid samples corresponding to the first target tracking angle, f represents the sampling frequency of the eye movement tracking device, and t represents the total sampling duration corresponding to the first target tracking angle.

[0137] In this embodiment, the second calculation module 30 is specifically used for:

[0138] Calculate the accuracy of each tracking angle in the original eye movement data, and perform a normal transformation process on the accuracy to obtain the transformed accuracy of each tracking angle;

[0139] Calculate the critical value according to the transformed accuracy of all tracking angles;

[0140] Traverse each tracking angle in the original eye movement data in descending order of the absolute value of the tracking angle. When the converted accuracy of the second tracking angle is less than or equal to the critical value, and the converted accuracies of all tracking angles greater than the second tracking angle are greater than the critical value, then determine that the second tracking angle is the partition angle of the medium and high recognition areas of the eye movement tracking of the eye movement tracking device in the measured direction, and the second tracking angle is less than the first tracking angle.

[0141] In this embodiment, the second calculation module 30 is specifically configured to:

[0142] For the second target tracking angle, calculate the included angle formed by the set visual stimulus point, the fixation point detected by the eye movement tracking device, and the pupil center point, and calculate the accuracy of the second target tracking angle according to the included angle;

[0143] Use the Boxcox method to perform normal conversion processing on the accuracy of the second target tracking angle to obtain the converted accuracy of the second target tracking angle.

[0144] In this embodiment, the second calculation module 30 is specifically configured to calculate the accuracy of the second target tracking angle using the following formula:

[0145]

[0146]

[0147] Where, A represents the accuracy of the second target tracking angle, n represents the total number of samples, i represents the i-th moment, α i represents the included angle, abs represents the absolute value operation, represents the line-of-sight vector formed by the visual stimulus point on the display screen of the eye movement tracking device and the pupil center of the artificial eye at the i-th moment, is the line-of-sight vector detected by the eye movement tracking device at the i-th moment.

[0148] In this embodiment, the second calculation module 30 is specifically configured to calculate the critical value using the following formula:

[0149] L limit = mean - k * SD

[0150] Where, L limit represents the critical value, mean represents the mean of the converted accuracies of all tracking angles, SD represents the standard deviation of the converted accuracies of all tracking angles, and k represents the correction coefficient.

[0151] In this embodiment, the control acquisition module 10 is specifically configured to:

[0152] When the artificial eye rotates to the target tracking angle and the next two consecutive angles, and the eye movement tracking device fails to collect eye movement data at these three consecutive angles, that is, the data loss rate reaches 100%, stop rotating the artificial eye and collecting data at this time. The angle immediately preceding the target tracking angle is the maximum angle that the eye movement tracking device can track.

[0153] According to the tracking angle range evaluation system of the eye movement tracking device provided in this embodiment, first control the artificial eye to continuously rotate with the center point as the starting measurement point and along the preset measurement direction at a preset angle gradient value. The raw eye movement data can be collected by the eye movement tracking device. Then, determine the partition angles of the medium and low recognition areas of the eye movement tracking in the corresponding measurement direction by the data loss rate, and determine the partition angles of the medium and high recognition areas of the eye movement tracking in the corresponding measurement direction according to the accuracy, realizing the calibration of the tracking range. And it is applicable to the head-mounted eye movement tracking device. The present invention can provide support for subsequent eye movement interaction applications.

[0154] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

[0155] More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection portion (electronic device) having one or more wirings, a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other appropriate processing as necessary, and then stored in a computer memory.

[0156] It should be understood that various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or a combination of the following technologies well known in the art can be used: discrete logic circuits with logic gate circuits for implementing logical functions on data signals, application specific integrated circuits with appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0157] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0158] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A method for evaluating the tracking angle range of an eye movement tracking device, characterized in that, The eye movement tracking device is internally provided with an artificial eye, and the method includes: Controlling the artificial eye to continuously rotate with the center point as the starting measurement point along a preset measurement direction at a preset angular gradient value, collecting raw eye movement data through the eye movement tracking device, and stopping rotating the artificial eye and collecting data when the data loss rate reaches a preset value; Calculating the data loss rate of each tracking angle in the raw eye movement data, and determining the partition angles of the medium and low recognition areas in the eye movement tracking of the eye movement tracking device in the corresponding measurement direction according to the data loss rate; Calculating the accuracy of each tracking angle in the raw eye movement data, and determining the partition angles of the medium and high recognition areas in the eye movement tracking of the eye movement tracking device in the corresponding measurement direction according to the accuracy.

2. The evaluation method for the tracking angle range of the eye movement tracking device according to claim 1, characterized in that The step of calculating the data loss rate of each tracking angle in the raw eye movement data and determining the partition angles of the medium and low recognition areas in the eye movement tracking of the eye movement tracking device in the corresponding measurement direction specifically includes: Traversing each tracking angle in the raw eye movement data in the order of the absolute value of the tracking angle from large to small. When the data loss rate of the first tracking angle is less than or equal to the preset loss rate, and the data loss rate of the next tracking angle greater than the first tracking angle is greater than the preset loss rate, then determine the first tracking angle as the partition angle of the medium and low recognition areas in the eye movement tracking of the eye movement tracking device in the measured direction.

3. The method for evaluating the tracking angle range of the eye movement tracking device according to claim 1 or 2, characterized in that, In the step of calculating the data loss rate of each tracking angle in the raw eye movement data, for the first target tracking angle, the following formula is used to calculate the data loss rate of the first target tracking angle: Loss=N 无效 / (f*t); where Loss represents the data loss rate of the first target tracking angle, N 无效 represents the number of invalid samples corresponding to the first target tracking angle, f represents the sampling frequency of the eye movement tracking device, and t represents the total sampling duration corresponding to the first target tracking angle.

4. The method for evaluating the tracking angle range of the eye movement tracking device according to claim 2, wherein The step of calculating the accuracy of each tracking angle in the raw eye movement data and determining the partition angles of the medium and high recognition areas in the eye movement tracking of the eye movement tracking device in the corresponding measurement direction specifically includes: Calculating the accuracy of each tracking angle in the raw eye movement data, and performing a normal transformation process on the accuracy to obtain the transformed accuracy of each tracking angle; Calculating a critical value according to the transformed accuracy of all tracking angles; Traversing each tracking angle in the raw eye movement data in the order of the absolute value of the tracking angle from large to small. When the transformed accuracy of the second tracking angle is less than or equal to the critical value, and the transformed accuracy of all tracking angles greater than the second tracking angle is greater than the critical value, then determine the second tracking angle as the partition angle of the medium and high recognition areas in the eye movement tracking of the eye movement tracking device in the measured direction, and the second tracking angle is less than the first tracking angle.

5. The method for evaluating the tracking angle range of the eye movement tracking device according to claim 4, characterized in that The step of calculating the accuracy of each tracking angle in the raw eye movement data and performing a normal transformation process on the accuracy to obtain the transformed accuracy of each tracking angle specifically includes: For the second target tracking angle, calculating the included angle formed by the set visual stimulus point and the fixation point detected by the eye movement tracking device and the center point of the pupil, and calculating the accuracy of the second target tracking angle according to the included angle. Use the Boxcox method to perform a normal transformation on the accuracy of the second target tracking angle to obtain the transformed accuracy of the second target tracking angle.

6. The method for evaluating the tracking angle range of the eye movement tracking device according to claim 5, characterized in that, In the step of calculating the included angle formed by the set visual stimulus point and the fixation point detected by the eye tracking device with the center point of the pupil for the second target tracking angle, and calculating the accuracy of the second target tracking angle based on the included angle, the following formula is used to calculate the accuracy of the second target tracking angle: Where A represents the accuracy of the second target tracking angle, n represents the total number of samples, i represents the i-th moment, and α i represents the included angle, abs represents the absolute value operation, represents the line-of-sight vector formed by the visual stimulus point on the display screen of the eye tracking device and the pupil center of the artificial eye at the i-th moment, is the line-of-sight vector detected by the eye tracking device at the i-th moment.

7. The evaluation method for the tracking angle range of the eye movement tracking device according to claim 5, characterized in that, In the step of calculating the critical value based on the transformed accuracy of all tracking angles, the following formula is used to calculate the critical value: L limit = mean - k * SD Among them, L limit represents the critical value, mean represents the mean of the converted accuracies of all tracking angles, SD represents the standard deviation of the converted accuracies of all tracking angles, and k represents the correction coefficient.

8. The method for evaluating the tracking angle range of the eye movement tracking device according to claim 1, characterized in that, When the data loss rate reaches the preset value, the steps of stopping the rotation of the artificial eye and collecting data specifically include: When the artificial eye rotates to the target tracking angle and the next two consecutive angles, and the eye tracking device fails to collect eye movement data at these three consecutive angles, that is, the data loss rate reaches 100%, at this time, stop rotating the artificial eye and collecting data, and the angle before the target tracking angle is the maximum angle that the eye tracking device can track.

9. A tracking angle range evaluation system for an eye movement tracking device, characterized in that, The eye tracking device is provided with an artificial eye, and the system includes: A control acquisition module, configured to control the artificial eye to continuously rotate along a preset measurement direction with a center point as the starting measurement point and a preset angle gradient value, collect raw eye movement data through the eye tracking device, and stop rotating the artificial eye and collecting data when the data loss rate reaches the preset value; A first calculation module, configured to calculate the data loss rate of each tracking angle in the raw eye movement data, and determine the partition angles of the medium and low recognition areas of eye movement tracking in the corresponding measurement direction of the eye tracking device according to the data loss rate; A second calculation module, configured to calculate the accuracy of each tracking angle in the raw eye movement data, and determine the partition angles of the medium and high recognition areas of eye movement tracking in the corresponding measurement direction of the eye tracking device according to the accuracy.

10. The evaluation system for the tracking angle range of the eye movement tracking device according to claim 9, characterized in that The first calculation module is specifically used for: Traverse each tracking angle in the raw eye movement data in descending order of the absolute value of the tracking angle. When the data loss rate of the first tracking angle is less than or equal to the preset loss rate, and the data loss rate of the next tracking angle greater than the first tracking angle is greater than the preset loss rate, then determine the first tracking angle as the partition angle of the medium and low recognition areas of eye movement tracking of the eye tracking device in the measured direction.

Images