Hi, I need an object to be attached to the ground once a target is recognized, but I still need the image target to be functional, because I'll have objects attached to the ground and objects attached to the target.

Is this possible?

Has anybody figured out how to set this up?

Wondering the same....

Hi, sorry to reactivate an old thread but did this ever get fixed?

For fixed world origin - do you mean first target or device? 

@vdasilva I am interested in use case 2. I have attempted to make the image target a child of the ground plane stage, but this did not work. Attempting to map the content from the image target(s) to the user defined ground plane. Thanks.

Hi @vdasilva

I would also like to scan an image target and anchor it to the ground plane. Right now, it seems as though the image target content and the Ground Plane content are on two complete different planes. When I hold the phone about 2 feet away from image target, the GP content appears on top of the IT content. I would like for them both to be on the same plane so that they can interact with one another. Could you provide a more detailed explanation?

Hello，I want to use Ground Plane and Image Targets，First，I use  Ground Plane,When I switch to Image Target,The identified model is shake ,Why is that,Thank you all.

Hi @vdasilva

Is it possible to get an example of the "usecase 2"? 

I would love to scan an Imagetarget and then choose where to put the object using groundplane but i simply cant figure out how to do it. 

I hope you can help me.

Hi @vdasilva

Is it possible to get an example of the "usecase 2"? 

I would love to scan an Imagetarget and then choose where to put the object using groundplane but i simply cant figure out how to do it. 

I hope you can help me.

Hi @Jarileon,

We have a couple of different approaches to this sort of functionality - and the approach you use is going to be dependent on your use-case.

Use Case 1: You want your experience to start from an Image, but you want the content to feel anchored to the environment.

Approach: Use your Image target as expected, be sure to use Extended Tracking by Enabling Device Tracker in Vuforia Engine 7.2 or later

Use Case 2: You want the user to setup a Ground Plane and you want the content from an Image Target to move onto the Ground Plane. 

Approach: Allow the user to setup the Ground Plane and once the Image Target is detected, then you can have the content move between the two by GameObject parenting. The Coordinate systems should be the same, just be sure to setup Device Tracking in Vuforia Engine 7.2 - World Origin should remain fixed in space

Please let us know if you have a different use case.

@emorling

There are lots of reasons why you could be having Jitter - please provide more information so we can try to guide you to a solution.

What device are you using? are you using ARKit/ARCore? How big is your Image Target? As mentioned by medabit, a video would be really helpful for us to diagnose any potential problems.

Would it be possible to get an answer from Vuforia? This thread has 20000+ views, so there seems to be a high demand for this. Please confirm if it at all possible to combine ground plane with image targets in a way, that you could position AR content with image targets on the ground plane? 

To me it seems that they do not operate in the same coordinate system - when an IT is found, it is positioned at (0,0,0) but simply setting the position of the AR content to the same point does place it on the target and keep the perspective correct. 

It would be great if you could explain the relation between IT and GP spaces.

None of the solutions in this thread gave the desired result (mostly because of the outdated code)

Hi. I also need to somehow be able to combine Image targets and Ground plane. Is this something that Vuforia now is supporting? I need to re-position content placed on ground plane whenever image target is visible. At least it seems to be really difficult, since it seems that ground plane and image targets have their own coordinate space - the Image targets origin seems to be in another place than the GP:s origin. 

Has anyone been able to figure out how to get Image target´s position on ground plane's coordinate system?

Hello @emorling,

Difficult to understand the issue with the provided description. If a video of the behavior can be provided, then it will give me an opportunity to analyze and replicate.

Thanks,

Vuforia Engine Support

@Medabit any input on this?

For this flow the new updates don't help:

1. Detect image target

2. Start extended tracking

3. Whenever image target is detected the user will get a jittery experience  (as the tracking gets re-anchored)

Ideally the image target would only initialize the extended tracking and after that not be involved again.

any ideas?

Did you find a solution to this?

I don’t think the current solution is satisfactory.

1. Image target is detected 

2. Extended tracking kicks in 

3. Image target is detected again AND everything is repositioned using the image target which provides a JITTERY experience 

please advise

Hey Folks,

The latest version of Vuforia (7.2) has made a lot of improvements in the area of using ARKit/ARCore for Extended Tracking (through the Positional Device Tracker).

If I understood the use cases mentioned on this thread, developers wanted to initiate an experience using an Image Target and keep it anchored in Place using ARKit. This functionality should work out of the box. Please refer to this doc for more info: https://library.vuforia.com/content/vuforia-library/en/articles/Training/Extended-Tracking.html

Please keep the feedback coming!

Vinny

augminded wrote:

I had to abandon Vuforia and use ARKit throught the Unity plugin. Works really nice. I recommend this approach. And with ARKit 1.5 and 2 I don't see anything in Vuforia that makes me wanna come back. 

Finding myself doing the same, it's a bit of an oversight on Vuforia's part - we want to include ground plane in a different scene to a scene with the marker tracking and the two do not work together in the same app at all because the ground plane requires a different device tracking mode to marker tracking.

I had to abandon Vuforia and use ARKit throught the Unity plugin. Works really nice. I recommend this approach. And with ARKit 1.5 and 2 I don't see anything in Vuforia that makes me wanna come back. 

So, what is the current status on this? Are any of the below solutions viable or is the functionality coming to Vuforia soon?

Hi,

I couldn't find any solution to this.

I have scenes where the world centers need to be FIRST TARGET and DEVICE TRACKING.

So, it's all mixed, but the FIRST TARGET behavior gets replaced with DEVICE TRACKING.

Would you please tell us how you fixed this issue? 

Thanks.

scott_a_ashton wrote:

Hi vdasilva,

Thankyou for the much needed update.  

Note: I wrote a long post here but I've deleted it.  I'm pleased to say I've got image targets and groundplane working nicely together in 2018.1f2. What took me hours and hours of pain and hair pulling on an older version of Unity/Vuforia now seems to work pretty easily. Nuff said I guess.

Cheers

If you share your solution with the community, we'll send you some swag! Ping me if you do!

gemelo wrote:

but can you give us an idea, at what time this will be?

We're looking to get it into our summer release, but no promises yet!

Hi vdasilva,

Thankyou for the much needed update.  

Note: I wrote a long post here but I've deleted it.  I'm pleased to say I've got image targets and groundplane working nicely together in 2018.1f2. What took me hours and hours of pain and hair pulling on an older version of Unity/Vuforia now seems to work pretty easily. Nuff said I guess.

Cheers

Thanks for the update. Definitely something to look forward to!. I think it'll be best if the function can be applied to all targets (Image, Model, Cylinder etc...). I believed that many people will start to ask for this because if it works well, it'll unlock all kinds of possibilities :).

Thanks vdasilva!

vdasilva wrote:

[...] We cannot wait to see the awesome experiences that you all build with the next version of Vuforia.

Thank you!

Thank you, sounds very good and I'm looking forward for that... but can you give us an idea, at what time this will be?

Regards

Thies

Hi Everyone,

We wanted to provide the community some updates on this thread. We understand this topic is something our developers are very passionate about. We have some updates that we'd like to share with the community.

The community outlined some use-cases looking to bridge Image Targets and Ground Plane.  Some time ago, I provided some alternative Unity components to aid in these requests in a Gist. While this solved some of the use-cases, it did not cover all the cases. I am currently working on an another set of components to cover the remaining cases. I'll update this thread once I have something to share. I’d like to point out that this approach does not require any updates to the Vuforia Engine – it simply leverages the Vuforia Engine in different ways.

There was also some confusion in regards to Vuforia's support for ARCore. As of Vuforia 7.0/7.1 - Vuforia does not  leverage ARCore.  In Vuforia 7.0,/7.1, developers targeting Android use Vuforia's own VIO instead of ARCore. As a result, developers may have of seen instability issues due to lack of ARCore. Internal support for ARCore is scheduled to be released in the next major version of Vuforia. 

Last year we announced Vuforia Fusion. We outlined a long-term vision of how Vuforia would work with platform tech such as ARKit and ARCore.  When we added support for ARKit, we took our first step towards Vuforia Fusion.  In the next release of Vuforia, we are taking another big step towards Vuforia Fusion. Developers can expect to see changes related to Extended Tracking and Device Tracking. These changes will allow Vuforia targets to leverage ARKit and ARCore as necessary. For instance, Vuforia will use ARKit/ARCore for targets that are "extended tracking". 

These upcoming changes should cover a lot of the use-cases outlined in this thread.

The team has been working hard trying to get all this functionality into the next release. We cannot wait to see the awesome experiences that you all build with the next version of Vuforia.

Thank you!

annafuste wrote:

I guess it makes a difference here depending on if you are using ARCore or ARKit. I have the exact same code for both but it will only work on iOS.

This is currently expected behavior as Vufoira 7.0 / 7.1 does not support ARCore yet.

Hey guys! Have any of you managed to get this going? Could anyone maybe share a unity project / working code samples?

I would be insanely grateful.

Thanks!

+1 request for help on this.

Hi,

I'm trying the code from Esnho, but it seems that latest Vuoforia release breaks something and I can't get the transform of the created anchor.

Have anyone made any more progress so far?

I'm working with Unity 2018.1b13 and Vuforia 7.1.31

Thanks in advance,

Alfred

Hi annfuste,

I tried Esnho's code too and it works well (I'm impressed and appreciative of the effort) but there's a initial drift of the 3D object. Did you encounter similar behaviour?

How about the rest?

Best regards,
Benjamin

Hi,

here's my progress so far. It may help some of you.

My goal is to try and anchor MultiTargets to the Ground Plane so I don't loose their position if I move away. 

I had been trying for long with a Pixel1. Vuforia was doing the Ground Plane over ARCore. It never worked and I am still trying to figure this out. If I anchor them to the Ground Plane, the anchored content always moves away.

BUT, I tried with an iPhone and it works now. I am using the code from Esnho for the TrackableEventHandler.

I guess it makes a difference here depending on if you are using ARCore or ARKit. I have the exact same code for both but it will only work on iOS.

Also, make sure that your targets are set to the right scale on the Target Manager. 1 unit = 1 meter.

I had mines wrong and this may cause issues when placing in space.

I had more luck using SetParent(transform) instead of transform.parent = transform but there is still a fair bit of drift

In this video, he did it, but i don't now how : https://www.youtube.com/watch?v=rcGQzuUwSA4

Hi Vinny, 

thanks for the solution.

It looks like there is a considerable offset when moving around compared to placing the content on the Ground Plane Stage straight away.

Do you have any idea what the problem could be?

I am trying with a MultiTarget instead of an Image Target. But I guess it should be the same.

Thanks!

I'd really like this too. My goal is to have Vuforia discover the Image Target and show the augmentation as usual and then switch to Ground Plane mode so that the user experiences more robust tracking and freedom of movement... Ideally there might be an interval you can set where Vuforia tries to discover the image target again and if it finds it updates the Ground Plane anchor for accuracy...

laurencenairne wrote:

- Ground Plane anchoring is still reliant on user interaction (tap)

This was intended. If this is a feature you'd like, I can look at a replacement for the AnchorInputListenerBehaviour.

laurencenairne wrote:

- Scale of content is fixed at the values set when the parent change occurs (i.e. after change, moving towards content makes it appear as though it is moving away at the same speed)

That sounds like an issue with a bad anchor. Are you seeing this same behavior when you place content onto the ground plane by itself?

A good start, but a couple of issues with it:

- Ground Plane anchoring is still reliant on user interaction (tap)

- Scale of content is fixed at the values set when the parent change occurs (i.e. after change, moving towards content makes it appear as though it is moving away at the same speed)

Hey Folks,

Here's a quick script that will take the contents of an Image Target and places it onto a Ground Plane:

https://gist.github.com/vad710/015629ca8d56b4a016e072cd0c2dd18c

HTH

Vinny

Just leaving this here for a solution :)

Apologies - I'm under a bit of time pressure here.

I'll try and post something more detailed in the next 24 hours.  

Hi scott_a_ashton, I'm very thanks for your sharing. I tried to get Image Target screen point by WorldToScreenPoint, but the value was pretty strange when I put Image Target and Ground Plane Stage on the same layer.. 

Here want to try about using the screen point of image target on the hit test of plane finder, which is INTERACTIVE mode. Is this a right way? 

It will be very thankful if you can explain more detail with your solution. 

Best Regards.

Hey scott_a_ashton,

Would you mind explaining a little bit more of how you got this working?

We are helping a professor try to do pretty much the same thing you are.

If you could even provide a small example or snippet I would be super grateful!

The final solution ended up doing a ScreenRay cast of the ImageTracker's screen position with the Ground Stage plane.  I did not make the ImageTracker a child of the ground stage.  Use the orientation of the imageTarget and the intersection position to set the transform of your (image tracker)content.  The results are pretty good.  The content is a child of GroundStage not eh imagetracker.  We only need the image tracker for orientation and (screen)position.

If you don't do the raycast, the final position is completely askew.  It looks fine from the position you locked it, but then you move around..... You have to wonder if the two systems are actually in the same space?

This will do till Vuforia sort themselves out!  

Hi Vuforia,

Do we have an update on this?

Like others here, the I was expecting that Fusion would mean that the 2 technologies would be used together.  In fact, at a conference (not that long ago), I asked the Vuforia speaker this exact question - and he said that it IS possible.

In fact I've had it working - but for some reason I'm unable to replicate my old results (on an older Vuforia?).  Please enlighten us of the most robust way of getting these 2 technologies to work nicely together.  

I for one would be hugely grateful.

Thanks in advance

ImageTargetTrackable works well, however orientation is not as expected. How to handle orientation if I use the same source code for detection of object targets?

@rsultangareev Hi, thanks for the post I tired it but it doesn't; work very well I have a image target which has the mid air stage as a child and that has the augmented gameobject as its chid but it doesn't work well I also changed it so there is a image target with the midair stage as its child and the augmented game object as a child of the image target it looses tracking and flickers a lot? did this happen for u?