Does the pitool included in the Pimax Experience include the eye tracking support?

I notice that eye tracking requires 263 of pitool, i plan on trying the eye tracking and want to know if i can do that and test PE at the same time?

P.S maybe in the future you can select games with your eyes and wink to start lol :wink:

Good question, as Im using one PC specifically for eye tracking right now, which has 263. I will test if the eye tracking works with the PE versions of PiTool, and report later!

4 Likes

If it does will you be adding eye tracking game selection to PE?

wink to start game wink twice to quit?

:wink:

I have requested and Im currently waiting for the source code of the Unity calibration tool and SDK from 7Invensun. Im sure there is no UI Input event handler for the eye tracking yet, but I should be able to use a custom made gaze UI interaction (which Ive been testing out in the past) and connect the eye tracking position data to the gaze pointer. If all goes well, eye tracking interaction should be in beta within a few weeks or even earlier.

The difficult part is to implement the click event and dragging - and how its supposed to be triggered. Normal gaze interaction such as the ones used in all sorts of 3dof VR units and cardboard apps, is based on gaze time. Look at something for a second and the pointer will “load” until it clicks or grabs the object. The difficult part is to determine when and how to release the drag. I have to do some research to see what solution is the most convenient and user-friendly. After all, Eye tracking UI interaction is a little more complex than hand tracking.

3 Likes

Woah there, I was kidding. Careful or this will turn into scope creep the size of Star Citizen.

I think you should focus on making sure that PE is the best launcher for games and all of this extra stuff like hand tracking and eye tracking can come after you make your first release.

Believe me I architect software for a living, I know a thing or two about scope creep.

5 Likes

Haha you are right :slight_smile: There’s tons of things to be done that are more important than eye tracking interaction. Although, I think having the eye tracking calibration built in into VRExp is step one and quite an important one.

A proper calibration might be needed every day or several times a day (you never wear the headset 100% identical two times in a row) so being forced to take off headset, start calibration, put on headset, calibrate and then maybe also take off again to exit the calibration before you put on your headset and start a game is just a hassle and not practical.

I know you can instantly exit the calibration by jumping into SteamVR dashboard and exit application, but many people might not know about this. And the calibration actually dont need SteamVR after all, so its not necessary to use it as a SteamVR application.

3 Likes

I think you have to really understand how seriously important drowhunters comment about scope creep is - it is the difference between a successful or failed project, and as I consider PE to be a major requirement for Pimax to jump out of it‘s super high maintenance device niche for die-hard VR nerds to reach the high maintenance for advanced VR users bigger niche (because let‘s be honest, it will never go past that one), the success of PE is rather important for Pimax.

A rugged tool reliably performing its core tasks is so much more important than adding eye-candy features like gaze-UI etc. if it by any chance compromises the functionality of the entire tool, or even just diverts resources away.

Why don‘t you just leave this one to be modded by others once you have made the source available ? If it works, cool; if it doesn‘t, it won‘t be associated with Pimax and the base PE tool Pimax release works well.

5 Likes

You definitely have a good point there, thanks!

4 Likes

Yes I agree if calibration needs to be done each time that would be annoying, but this shouldn’t be a priority for initial release is what i’m saying.

Hopefully hamdtracking is on a separate branch as I wouldn’t consider this a priority feature(i dont even have a leop motion)

I know how it is you are having fun and it is easy to get lost down the rabbit hole.

A. The eye tracking doesnt even work well yet
B. The % of users who actually have eye tracking is small
C. @Axacuatl said it’s way more imortant to focus on a core release. The ONLY thing important right now is making a mostly bug free PE that optimizes and launches games.

What we usually do at my work is we call a feature freeze on the product and only work on fixing bugs before doing a release. Any thing else goes onto a backlog and then once the release is put out . say v.1.0.0 you can then work on adding one or two of those new features and plan them for the next release.
Usually you will operate on 2 week sprints where you plan to only get done what you think you can do in that time before doing another release. rinse and repeat.

betweek releases bugs will be reported and you can put out interem patches increasing the version number.

Its called semantic versioning where the number signifies the importance of the change

so 1.2.3 = MAJOR.MINOR.Revision

Major number increase is usually reserved for major changes , usually not backwards compatible.
Minor is non-breaking changes
Revision is for tiny changes that wont break anything.

This all usually mangaged by having different branches in github.

All of which you can do once you open source.

Sorry i’ve rambled too long :smiley:

4 Likes

Just to add justification to their point: I have no intention of using eye tracking until forced to, so the time that you spend on this would be “wasted” from my perspective.
-Keep in mind that I don’t actually think that you would be wasting your time, I’m just providing another voice.
Awesome work, dude!!! No sleep for you for the next 2 years! :stuck_out_tongue_winking_eye:

2 Likes

Any update on this? I just installed eyetracking last night I guess ill find out later if it works or not. I was too tired after cutting up my 5k+ to install it to try it :roll_eyes: