imarin18
-
Posts
50 -
Joined
-
Last visited
Content Type
Profiles
Forums
Blogs
Downloads
Posts posted by imarin18
-
-
On 12/3/2019 at 5:14 PM, Daniel_Y said:
Thanks for your feedback. We noticed this issue and will be addressed in the next release.
Dear @Daniel_Y,
I updated SR Runtime version to SRanipal Runtime (1.3.2.0) and checked this issue. I measured the same parameters as before, thought it seems that the device outputs different pupil data, whereas the time stamp is same. Please refer to the attached my data.
Could you please share the update regarding this issue?
Thank you very much for the information you could provide.
Best regards,
imarin18
-
Dear @maple
I suppose that you faced the situation possibly due to the following two reasons.
1: Your computer does not have enough specification to get the data at 120Hz sampling rate. A computer with better specification could enable it to sample at 120Hz.
2. If you look at only the initial samples (i.e. first 30 seconds after you start sampling), you will see lower sampling rate than 120Hz.
The following article may be helpful for you to understand these points.
https://doi.org/10.3389/fpsyt.2020.572938
Best regards,
imarin18
-
Dear @prajaktakhandve
I'm sorry for my late response. I could not get the notification from this thread.
I recommend referring to the following research paper: https://doi.org/10.3389/fpsyt.2020.572938. The authors open the source code on GitHub: https://github.com/MotorControlLearning/SaccadeVR-mobile.
Best regards,
imarin18
-
Dear @jboss
Thank you very much for your response.
I think your understanding is correct. We need to remember that Unity playing does not always work along with eye tracking. In your script posted previously, you wrote the code below. This stops only the playing of Unity, not eye tracking. Thus, if you want to stop the eye tracking in parallel, one of the workarounds would be to use UnixTime (i.e. SaccadeEndTime) as a trigger to stop the eye tracking, as designed in the GitHub page: https://github.com/MotorControlLearning/SaccadeVR-mobile.
if (Application.isPlaying) { Application.Quit(); }
Best regards,
imarin18
-
Dear @jboss
The reason why you face the frozen Unity is that the eye tracker is still measureing the eye movements, while the process in void Update () is completed. As long as you continue to record the data of eye movements, Unity playing will not end. The GitHub page says as below.
"Please be noted that you will see Unity seem to be frozen if you click the stop button on Unity for a specific case. Unity seems to be frozen in this situation because the sampling of eye tracking is still working on the background. As explained in the following algorithm flow chart, the sampling of eye tracking works independently from Unity. The playing of Unity is stopped automatically when all the saccade tasks are completed. We set the maximum sampling number of eye tracking to 120 Hz * 1800 seconds in line# 175: private const int maxframe_count = 120 * 1800; of C# script: Saccade_measure_rev1.cs. Therefore, if you click the stop button on Unity within maxframe_count time range, Unity seems frozen (actually not because the system still waits for the sampling of eye tracking to finish). On the other hand, if you click the stop button beyond maxframe_count time range, the playing of Unity stops immediately. The maximum number of maxframe_count can be adjusted accordingly."
If you adjust the value of private const int maxframe_count on the script according to your preference, I suppose you can avoid the frozen Unity.
Best regards,
imarin18 -
Dear @jboss
That's great to know you were able to confirm the sampling interval properly after changing the laptop. I suppose that the eye tracking will run at 120Hz, even if you change the FPS since the eye tracking should work independently from Unity fps configuration.
Best regards,
imarin18 -
Dear @jboss
What measurement parameter did you check for the interval? As posted in another thread, time_stamp parameter recorded by SRanipal SDK does not provide the value properly (HTC also recognises the issue). Otherwise, it would be better for you to try with another laptop that meets the requirements recommended by HTC (https://developer.vive.com/resources/vive-sense/hardware-guide/vive-pro-eye-specs-user-guide/).
Best regards,
imarin18 -
Dear @jboss
As the paper reported, the longer sampling interval of 16/17 ms would be due to the clock timing of your laptop. Or if you evaluated the sampling interval for the initial 30 seconds, it would be better to check the data over 30 seconds after you start the recording.
According to the script provided in the paper, the recording continues in a while loop for a specific number of samples (see the line # 175) regardless of whether you click Play or not. Thus, the editor may have been frozen because the while loop was not completed.
Best regards,
imarin18 -
Dear @prajaktakhandve
As people have mentioned already, Unity FPS and sampling frequency of eye tracking should be handled and considered separtely. The eye callback function that HTC provides enables you to sample eye data at 120Hz, while FPS changes depending on the VR environment you design. It seems the following article will be of a good help for you.
https://www.frontiersin.org/articles/10.3389/fpsyt.2020.572938/full
-
1
-
-
Dear @Tony PH Lin
Thank you very much for your check and the information. It is really helpful.
Best regards,
imarin18
-
1
-
-
Dear HTC team,
I'm writing to ask you about the specification of HTC VIVE Pro Eye. The field of view is 110 degrees according to your information. Is 110 degrees for both vertical and horizontal or diagonal?
Thank you very much for the information you could provide.
Best regards,
imarin18
-
As @Corvus mentioned, we need to see Unity FPS and eye sampling seprately. These two parameters: frame rate and sampling frequency of eye tracker are indepedent from each other. For example, while the Unity FPS changes depending on the VR design and the computer specification (e.g. sometimes it is 50FPS and it changes to 100FPS at another time), the sampling frequency of eye tracker is basically consistent at a specific value. If you use the callback fucntion, the sampling frequency should be more or less at around 120Hz, whereas the Unity FPS may fluctuate. Specifically, while Unity Update() is running on one thread, Eye call back funciton is running on another thread.
However, as I wrote in another thread (
), it seems that the current SRanipal SDK does not properly output the time stamp data. Thus, it may be difficult to evaluate the sampling interval or frequency of eye tracker correctly, using the time stamp data. I instead used DateTime.Now.Ticks property on C# and observed that the sampling frequency was more or less at around 120Hz.
I hope this helps.
Best regards,
imarin18
-
Dear @Daniel_Y
Thank you very much for your reply and the explanation. I understood.
Best regards,
imarin18 -
Dear @Daniel_Y
Thank you very much for your reply and the explanation.
I understood. Is my understanding correct for the calculation of angle in degrees from the data of gaze vector, following the trigonometric functions?
Best regards,
imarin18 -
Dear @Daniel_Y
I think I figured out the calculation of angle in degrees. We simply need to follow the concept of trigonometric functions, don't we? For example, if we want to calculate the angle of our eye movements in X direction, we just need to get the arc tangent from the data in X and Z directions.
Could you please explain what 0 and 1 mean in the normalised pupil position in terms of angle in degrees?
Best regards,
imarin18
-
Dear @Daniel_Y
Thank you very much for your reply and the information. First of all, I made a wrong description in my previous post. I was supposed to write "Data in degrees assuming that the normalised data are between -pi and pi.", not "Data in degrees assuming that the normalised data are between -pi/2 and pi/2."
Could you please explain how the device normalises the gaze direction data? For example, what is the situation when the gaze direction on x-axis is 1?
According to the technical specification, the trackable field of view of this headset is 110 degrees. I reckon that -1 of the gaze direction data means -55 degrees and 1 means 55 degrees. Is my understanding correct?
I check the pupil position as well. Could you please explain the configuration of pupil position as well? Does 0 mean -55 degrees and 1 mean 55 degrees?
Thank you very much for the information you could provide.
Best regards,
imarin18
-
Dear @Daniel_Y
Thank you very much for your reply and the information.
According to the data from my measurement, the gaze direction data are not in the range between 0 and 1 as shown in the figure below. The data even have negative values. In my measurement, I intentionally moved my eyes at +/-8 degrees from the central point. Assuming that the data were normalised between -pi/2 and pi/2, I calculated the degrees and found the subsequent figure, showing that my eyes moved to +/-8 degrees as expected.
I'm sorry to ask you again and I may be still wrong, but could you please double confirm the configuration?
Normalised data in my measurement.
Data in degrees assuming that the normalised data are between -pi/2 and pi/2.
As for the normalised pupil position, do you mean the position from the direction of the VR headset?
Thank you very much for your support and the information you could provide.
Best regards,
imarin18
-
Dear @alwyuyang
Just to clarify again, I am not sure if my understanding is correct. According to your result, it looks that the validity value outputs 8 despite that you don't have any values for the measurement parameters. Have you tried to get the data while you are closeing your eyes completely and look at the validity data?
My personal opinion is that it would be better to use only the data with 31 on the validity data.
Best regards,
imarin18
-
Dear @alwyuyang
I suppose that the technical support team meant that the huge gap from 62613 to 4205382 in timestamp data would be also a part of improvement.
I'm not sure exactly about what you meant by "when it failed to measure", but when you get 8 in your validity data, I reckon that you have the valid data only for eye openness (SINGLE_EYE_DATA_EYE_OPENNESS_VALIDITY) having the value of 8 and the other four parameters have the value of 0, meaning that the four are not valid.
Best regards,
imarin18
-
Dear @alwyuyang
As explained in the following post, the timestamp cannot be correctly measured with the current SDK version according to the technical support team.
If I understand correctly, I reckon that the value of validity follows the table below. If you sum all the decimal digits (1+2+4+8+16), you get 31. Though again I am not sure if my understanding is correct.
Enumerator
Binary digit if valid
Decimal digit if valid
SINGLE_EYE_DATA_GAZE_ORIGIN_VALIDITY
00001
1
SINGLE_EYE_DATA_GAZE_DIRECTION_VALIDITY
00010
2
SINGLE_EYE_DATA_PUPIL_DIAMETER_VALIDITY
00100
4
SINGLE_EYE_DATA_EYE_OPENNESS_VALIDITY
01000
8
SINGLE_EYE_DATA_PUPIL_POSITION_IN_SENSOR_AREA_VALIDITY
10000
16
Best regards,
imarin18
-
Dear HTC support team,
According to my investigation, it does not seem that the gaze direction data are normalised between 0 and 1 as described in the manual. I reckon the data are in the range of [-pi/2 pi/2], though I am not sure if my understanding is correct. Could you please confirm the data configuration?
Best regards,
imarin18
-
Dear @alwyuyang
As in the following thread, the combined data are not supported at this point.
Best regards,
imarin18
-
Dear @alwyuyang
I suppose you can get the validity data from "eye_data_validata_bit_mask" in your code. The type of the output is UInt64, not bool.
Best regards,
imarin18
-
Dear @kbombeke
I observed the similar situation previously. I recommend using the information of eye validity first.
If I understand correctly, the following shows how it works, though I am not sure if this is correct.
Enumerator
Binary digit if valid
Decimal digit if valid
SINGLE_EYE_DATA_GAZE_ORIGIN_VALIDITY
00001
1
SINGLE_EYE_DATA_GAZE_DIRECTION_VALIDITY
00010
2
SINGLE_EYE_DATA_PUPIL_DIAMETER_VALIDITY
00100
4
SINGLE_EYE_DATA_EYE_OPENNESS_VALIDITY
01000
8
SINGLE_EYE_DATA_PUPIL_POSITION_IN_SENSOR_AREA_VALIDITY
10000
16
I hope this helps.
Best regards,
imarin18
How are the data of time stamp and frame measured?
in VIVE Eye and Facial Tracking SDK
Posted