#š¤Æāaugmented-reality
1 messages Ā· Page 5 of 1
Oh, MARS itself
I am not sure on the answer myself, but if the question here goes unanswered, you're welcome to post to the unity discussion forums in order to make sure your question doesn't get lost and never answered on discord (the nature of discord being a scrolling feed of messages makes it easy to lose questions)
Yes, will do. Also, this script I posted is just to hide the vertical planes, of course is far from ideal but it works for my case... But thanks!
The Meta Simulator does not run when I hit the play button even the Meta XR Simulator button is enabled. The "Game" screen stays black and none application is launched in a another window or tab. I already changed to use the plugin OpenXR and lauched the synth server.
Having some issues with **OVRCameraRig ** (or at least I think this has something to do with the issue)... basically I've added an option to my in-game menu that lets the user toggle passthrough ON or OFF, which is just a simple command OVRManager.isInsightPassthroughEnabled = true;
I'm noticing that if I toggle it a few times, the FPS begins to dip and the objects in my scene begin to get very fuzzy and somewhat pixelated, especially on the Quest 2. I notice this really only happens in my production builds as well, when I run from the Unity editor this doesn't happen. Any ideas?
Some of my graphics settings include: ASTC compression format, normal quality lightmap encoding, low resolution shadows, hard shadows only, Late Latching (Vulkan), defaults for Graphics tiers, 2x Multi Sampling, Anistrophic forced on... I've tried toggling a bunch of different things but the prod build never seems to be the same quality as the development builds from the editor.
Dont you need the oculus xr plugin active?
I think you also need to enable it in the top toolbar
Best to update Unity, oculus plugin, and meta sdk.
If it still happens after that, check the profiler to see where the issue is and submit a bug report so it can be fixed
Thanks! In documentation is writen that I must choose between Oculus XR Plugin and OpenXR and the simulator only works with OpenXR. In the project "attention waring" point it as well. The Unity Editor Ć© updated and the SDK is relatively recent as well, I am using Meta All in One v67. I guess Meta does not have a good channel for communication and their forum Ć© quite abandoned š„²
I think that's the XRIT simulator and not the Meta one?
It's a bit all over the place now
On actually I do not know if the Meta one uses the XRIT under the hood, I must research more about it. I am pretty new one in this stuff of developing XR using Unity. I am using this documentation: https://github.com/Oculus-VR/homebrew-repo/blob/main/meta-xr-simulator.md#develop-openxr-app-using-unity. I am trying to know some other developers that know something about Meta SDK and MacOs being good friends but it is kind of a empty space between the other realities. š
Maybe check the official docs
https://developer.oculus.com/documentation/unity/xrsim-intro/
@teal elm
Meta XR Simulator is a lightweight XR runtime built for developers that enables the simulation of Meta Quest headsets and features on the API level.
Thanks, It is right this one documentation that I am following. If you go to the "Install Meta XR Simulator (macOS)" part of it documentation they will lead you to that link that I posted here before.
Ahh I missed the MacOS part. No experience there from me
Thank you so much for you attention, I will try to get some more information about it. Wish me luck!
Hello, i need help, i need unity subsystem and AR simulator in my project but i cant find them in the package manager
Is there anyway possible that I can connect my unity AR with react native-expo based environment?
Did you figure this out? I'm not entirely sure what you mean by unity subsystem and AR simulator
There may be a solution but that would be independent of it being an AR app, yes? Since the main problem is how to link the react app (if I'm understanding this correctly) to unity
Hi,
I have a nevigation scene and 5 AR scene and i want to switch between these scenes. But when I am doing this first scene run fine and when I click back and return to navigation scene and then move to same AR scene or any different AR scene error indicates that The object of type āCameraā has been destroyed but you are still trying to access it.
Is there any solution for this??
Using editor version 2022.3.3f and project AR mobile
hello, i try to get a stencil shader to work for hololens 2, but it isnt working perfectly.
i use the mrtk3 standard shader for it.
the problem is, depending on how far or close i am to the object it appears or disappears and so far i couldnt find which settings in the shader has to be enabled to get it to work correctly...
maybe someone can help?
Edit: nvm, solved the problem with using the graphic tools standard shader.
- Update unity, that version is old
1.1 also update ARF/ARCore - Share the code you use to switch scenes
- Which line is this error on?
how long does it take these days to put an app on teh quest store?
Yes, ar foundation does not support the mrtk3 standard shader, glad you got it figured out!
Depends on a lot of factors, but usually within a week for us. It could be a vendor status thing though...
Just be aware their rejections or change requests aren't usually emailed to you. So you need to keep an eye on the dashboard
sweet thanks for hte info
Anyone in here developing for AppleVisionPro? I have this weird issue when using the simulator, where logic is changing the material on my gameobject, but its only changing in the editor window, not the simulator window.
Hello, im new here xd. im a student creating our AR app for our capstone. I just wanna ask for help from yall seniors. Is it possible to make a merging system like for example theres 2 image the first one is a marker for a hydrogen atom, the second image is a marker for an oxygen atom. is it possible to mix them together to create a water molecule? im currently using vuforia engine and idk if it can support this idea
like you just swipe the 3d models of the molecules together and they mix together*
You can track the images and check their absolute world position, then use that to drive what logic you want
How do you change the material? Are you sure, the material is supported?
I did finally notice some material shader console errors. so i'll look into that
Guys I want to create an AR app .when I watch a video and it will pop up an AR object like in a particular scene it will pop up AR content
Dont be fooled by the simulator errors on unsupported materials tho. The simulator does not 1:1 simulate the behaviour as the actual device. But still, those errors could be an indicator, just heads up for testing on the real device over the simulator
Wheres the actual question?
Anyone in here worked with Personas on visionOS and how to show them in Unity? Right now, I can show them with webcamtexture support, but I am wondering if there is like a way to get the transparent one instead of those default backgrounds.
I am developing for the Apple Vision Pro. Porting an XR project. All these sphere's are XR Simple Interactables (and the capsules too). They are all set up the same. Very similiar heierarchys with each interactor. But, for some reason, I can't 'select' the ones with red 'x', but all the green checks are fine. If I double the collider size of the red 'x' ones, it works, but it seems like that just 'wakes' them up, because if out of playmode, i double the collider size, then play again, the selection doesn't work.
if i adjust the scale of the collider on the XR Simple Interactable at runtime, i'm able to interact with it. Smaller or larger. Doesn't matter.
if i disable/enable the collider component, or gameobject, i'm able to interact with it
are you sure, there are no like double components or child components probably blocking it? Did you test to create a fresh version of those "buttons" and see, if the issue persists?
We discovered the issue was something to do with the animator animating the scale of the objects. My guess is poly spatial didnāt know to update its links that the scale had changed. We discovered the difference between the interactors that were working, and the interactors that were not, the working ones as part of the reveal animation deactivated/activated the XR interact component. so we added those same keys to the interactors that were not working and that fixed it
Thanks for that insight. Id add my own inheritance of the xr interact then to just listen for the runtime events to recall the collider to activate them if you dont want to use animation keys. But might just be personal preference
Why is my script not running on my HoloLens when it works perfectly on my tablet? Does anyone know how to resolve the problem ?
I only see the first step, where it says that the program is being initialized (this step usually takes less than a second). Normally, I would be redirected to the next step, but in this case, I am stuck on the first step.
You def. should add some logs to your whole code to know, what you are actually doing on runtime
With Polyspatial, does anyone know of any examples showing multiple bounded volumes in the same app? even if its just someone showing it in a youtube video?
The polyspatial samples themself are showing this I think.
Unless I missed one, they are all a single bounded volume. But, the documentation says multiple is possible
You can have different scenes with different volume cameras using different volume settings
So, they would appear as more than one volume cube that can be moved around and placed by the AVP? In "shared" mode, your application coexists with any other applications that are active in the shared real-world space. Each application has one or more bounded volumes (see below), but no unbounded volumes. The position and orientation of these volumes (both relative and absolute) is opaque to the app. Input in this mode is limited to a ā3D touchā mechanism, via the PolySpatialTouchSpace device (see Input). In addition, ARKit information such as hand position, planes, or world mesh is unavailable in this mode.
If you mean the volume box, that you can move like any window on the AVP, then yes. If you use bounded volume, it can coexist with different apps
Hey guys, hope everyone's doing well!
I have the following case in AR and could not find any solution online. What I am trying to build is an inside building navigation and I have drawn a 3D model of the location that I want to navigate inside, however I am not able to get my drawing to fit 100% in the real world based on the user rotation. So my problem in short is how to get the user's rotation so I can get proper navigation. Thanks in advance!
if not 100%, how far is it off. Could it just be small tracking issues that are normal for AR or is it like totally off?
To all visionOS devs out there. Is there a Unity API to access the cameraframe in the latest visionos?
my problem is how can I get the drawn map to rotate according to the actual location based on the user's rotation, I don't know if I'm being clear. So my main issue is basically getting the user's rotation so I can fit my 3D map accordingly
the user rotation will always change, as the zero point of your virtual world will reset on every app start. You need to get some real world feature points to track to, otherwise it wont work. What platform are you developing for, ios, android, visionos?
Did anyone ever encounter this issue here? Seems like it happened before: MissingMethodException: Default constructor not found for type UnityEngine.XR.VisionOS.VisionOSHandProvider
the user rotation will always change, as
Seems like the issue was with IL2CPP packing and build resizing. There is an option on visionos settings to avoid this error, because it could not load the hand subsystem corretly otherwise and break the whole arsession
is Plane classification possible for ARCore and ARFoundation?
Greetings, would anyone be able to help me set up a basic AR system?
This is my first time working with AR, I've been trying to get something to work for weeks but I've had no success. No tutorial or forum I've followed has been of any help.
I would please just like help to come to understand how to work with it, and how to set up this system.
The system needs to do the following:
- It should be able to be turned off and on.
- Using the Mobile Phone's Camera, and it should look for a picture, and then project a 3D model onto where the picture is.
Please, any help would be greatly appreciated.
Hello, can someone please recommend a website where I can find free 3D models that I can use in AR? I need specific 3D models for my thesis and I have already checked the asset store and a bunch of other websites but with no luck. Any recommendations would really help. Thank you.
I have a 3D model for a Capybara that I am using as a placeholder for my above mentioned app
Thank you, the models I am looking for are the following: a trophy/award, badge, citation symbol, children, wedding rings, and something that represents scientific/academic publications. I can find all of them as 2d icons, but I am trying to utilize the 3D space as suggested by my supervisors.
In addition to those, i am also looking for 3d models for AR/VR glasses, chemistry-related, physics-related, pretty much a wide range of models, and I find out what I need as I go. Unfortunately, most models I find are paid so I have to look really hard to find free models.
i am unfortunately not that familiar with those types of models and where to find them. I am sorry.
In the worst case scenario, if they are not very expensive, then perhaps those paid methods you found could be worth it.
But if I find any such models, I will notify you
What you are probably looking for is called image tracking, the online tutorials is your best bet. !learn but for image recognition apps I use mindar or aframe, I don't like to run all that through unity unless there is a reason for it. Especially if it's a non game, because now you aren't allowed to use personal edition and subject to industry pricing.
:teacher: Unity Learn ā
Over 750 hours of free live and on-demand learning content for all levels of experience!
I see. Though the thing is that I have looked upa few online tutorials and followed them to the letter, but they just don't work.
Though, I'll give it another shot.
But for how I learn, if I can get a simple example of a working script for AR, I would learn from it.
I was never one to like to copy work, but I want to understand how it works and why, and I like to use examples to reverse engineer it, if that makes sense.
Unfortunately there are only a few options. Find them somewhere free, find them somewhere paid, pay someone to make them or make them on your own. My advice is to use blender, follow a few online low poly tutorials, and whip something together. You could also use just blocks and cubes but then you aren't learning as much as you could. There are also things like polyhaven, cgtrader and turbosquid if you don't mind poorly optimized or random quality free models and hunting for each one separately. It's a pretty specific list though so you'll probably need to pay at some point for it.
I don't have one that I can share, it's either connected to a client or one of our products. You could use the built in AR template but it's not very usable for new developers.
Indeed.
I have been tinkering with the template but it is very hard to understand or reverse engineer.
But thank you for your help either way, I appreciate it
Thank you
Anyone knows about this?
Anyone know of a tool I can use to access Apple EventKit inside Unity? To do things like Read/Write to the Calendar/Reminders for VisionOS?
Is it possible to build an app for AVP without a physical Mac? Using cloud mac and then TestFlight or something similar?
and if possible, is it actually feasible? or such a slow painful process that it's not worth it?
Context: I'm a Unity developer with a Windows desktop and I have a small APV project about to start, would rather not have to buy a Mac if I can avoid it.
Without a Mac, you're gonna have a hard time. Because:
- you need a Silicon Mac to even install the polyspatial packages.
- The simulator only works on Mac
In the past I've done iOS builds using Unity Cloud Build. And then Used a post-build script to push to AppStoreConnect. But, I don't think thats currently possible for VisionOS.
Someone suggested to try using a cloud rented silicon mac. You'd only need it to install the polyspatial packages, testing in simulator, and doing the build. The rest you could still work on in windows (is my understanding)
Yeah cloud rented silicon mac is what I was thinking about. Why would I need to use the simulator though? Isn't running in Unity editor enough?
Perhaps. But, the simulator is the best way to make sure you have your pinch inputs set up correctly.
hey @long glacier i see ur content on utubee occasionally, good turtorials mate!
TYTY
One day I'll do more. I missed doing a new one every week. I'm just so busy now.
Polyspatial/VisionOS question: Anyone know of one of the sample scenes, that on build, scaling of the bounded volume works? I'm using a Volume Camera and a Volume Camera Resize Listener. But, user is only able to move the volume. They can't resize it.
Polyspatial/VisionOS question: Anyone
i didnt get this , what do you mean by independent of it being an AR app? currently i m trying to build ar using mapbox and unity , but the thing is only "AR" part of the app is in unity and the rest of the app like the map features or ui or anything else is on react-native expo , so my question was is there any way possible i build that AR on the unity and export it or display in react native app?
if anyone else can also help me with this , it will be of great help
This is probably what you are looking for
https://unity.com/features/unity-as-a-library
To their point, where the gray area is, is how to connect that library with react native itself.
https://github.com/azesmway/react-native-unity
May help
THANKK youu ill check and let you know ā¤ļø
hey guys can i do augemented reality just with a phone???
That question is so vague. What do you mean?
I mean do i need an occulus or ar,vr headset
Hello guys
there is AR built in with most phones. Not sure we are talking about the same AR š AR is just digital content on top of real world image/camera feeds aligned by the motion sensors or image detection of the phone. Headsets are mostly VR or when using a decent passthrough, AR phone hardware with a headset strapped on
Yes, that is possible assuming the phone supports the features you need, which is dependent on what make and model it is, but basic AR is compatible with a majority of smart phones out now.
Thats just a default AR Foundation restriction though. There are also other frameworks built with unity (easyar, maxst, vuforia, opencv) that opens it to all devices. And other engines like aframe/mindar if you don't want to be restricted to unity or want to work in webgl. There is also a mindar for webgl for unity you can use.
There was also an OpenCV for AR Foundation plugin at some point you could use as a fall back.
But yeah, what Unity has built in by default is just arkit/core devices (some android and ios mobile devices).
Unity 6 (and some instances of 2022) have no soft shadows when XR is enabled. I would like to ask if people could vote on the issue to get some publicity, as it is hurting many projects. It happens with both ARCore and Oculus/Meta plugins.
Steps to reproduce: 1. Open the āIN-70214ā project 2. Ensure that Soft Shadows are enabled in Assets/Settings/URP Renderer asset ā I...
I am getting this error here and there seems to be no .1 version of vivox but the visionos package from backend is trying to get it:
Package com.unity.services.vivox-visionos@file:/Users/nsynkjohannes/Downloads/com.unity.visionos.vivox.unityVisionOsSdk-16.5.1.tgz has invalid dependencies or related test packages:
com.unity.services.vivox (dependency): Package [com.unity.services.vivox@16.5.1] cannot be found
hi guys, i went through the old texts but am still not clear what i should use for object decetion in unity for android(later ios and windows tablets). So far my option seems to be tensorlite or mediapipe as I want to detect a custom object and anchor its position if the real object is lost in the real envionemnet(using arcore or arkit later). Would this be a good approach to start? Unfortunately i have not worked with both mediapipe and tensorlite, is there something i should keep in mind or approach this differntly?
Looks like Unity is referring to vivox versions they do not even have, as the latest is 16.5.0 and not .1...
For anyone having issues. you can add the folder structure to your packages/ and just change to the .0 version for now until Unity fixes their dependencies
Hello everyone, I'm having trouble with the AR Raycast Manager after I change scene and then go back into an AR scene.
Autoconnected Player "Autoconnected Player" No active UnityEngine.XR.ARSubsystems.XRRaycastSubsystem is available. This feature is either not supported on the current platform, or you may need to enable a provider in Project Settings > XR Plug-in Management.
So, I'm assuming when changing scenes the AR session breaks and the XRRayCastSubsystem loses references to the MARS / AR Session?
I go into a UI Scene into an AR Scene (All works well), I go back into the UI Scene and then back into the AR Scene and the planes are being detected and all, but the AR Raycast is not working.
Anyone have any insights into this ? It's kind of driving me mad.
Also, anyone know how to deal with changing scenes in AR ? Like, how to properly dispose and re-start a AR Session, I feel like it breaks on any scene change.
For anyone suffering with this, this fixed it.
I am getting this error here and there
Hey so I have a problem in mixed reality multiplayer on the meta quest 3. I am trying to develop a mixed reality project with multiplayer using netcode for gameobjects on unity. The issue that I have right now is that since both headsets won't have the exact same boundary, the position of the objects won't be in the same location across headsets. I was wondering if anyone can shoot me in a direction to look for a solution or give some thoughts or inputs into this issue because honestly I am at a loss.
Take a look into Meta shared spatial anchors. I think Unity also made it work in a demo, but unsure
Hi everyone,
I'm currently working on an AR project using Unity with AR Foundation and ARCore, and I'm facing an issue where AR assets start drifting from their original positions when I move or rotate my phone. Even though the assets are parented to an anchor, they seem to lose their correct positioning and drift after some time or when the phoneās orientation changes.
Has anyone experienced this before, and are there any known solutions to prevent this drift?
A bit more context:
The assets are anchored using AR Anchors.
The project involves placing models based on real-world coordinates.
The drift occurs after I walk around or rotate the device, even though the environment has enough visual features for tracking.
Any advice or suggestions on how to improve anchor stability or minimize drift would be greatly appreciated!
Thanks in advance for your help!
Updating Unity to 6 and the latest ARCore seems to help some projects
I wonder if using 60fps helps, never tried that tho. Is performance stable in general?
Sadly ARcore depends a lot on the device and tends to drift a bit yes
If it's very reproducible you could file a bug report, but I am unsure if that will get through
Can I build a WebGL application using The ARCore Geospatial Creator for Unity?
I've been exploring the ARCore Geospatial Creator for Unity but am having trouble understanding the documentation. Is it possible to build WebGL applications with this package, or is it limited to Android platforms only?
No, ARCore is only android native
For WebAR I suggest other platforms like 8th wall/niantic studio or wonderland
There are some web AR tools for Unity as well, but I havent tried them
Thanks, I checked 8th wall and immersal, webAR + Visual Positioning System is usually a paid feature. And the pricing is a bit high
Niantic Studio just changed their pricing. Not sure how much it was, but I think it is better now. Still not cheap still
yeah doesn't look cheap still... I also found Adobe Aero pretty cool (based on the ARCore as well), but its annoying that user hast to open the app to view the experience
Hey, Using Mobile AR with Face Tracking, is it possible to render a mesh behind the human head?
Hello, I am trying to pinch to scale, when I searcher for how to do this, most results suggest using ARScaleInteractable, but when I added this component to my cube there was a warning that it is now deprecated and that I should use XR grab interactable. I am now using XR grab interactable, and I can transform position and rotate but I am unable to scale. I am unsure if I have the wrong settings used for "track scale" or if i should be interacting in a specific way for the pinching to be detected as a scale up or scale down. Any thoughts?
Here is a phot of my XR grab interactable:
hey there, i've been bashing my head into a wall trying to figure out why my project will build but not deploy to my device and i've hit a dead end
i've messed with api levels, clearing device cache, different devices, different unity editor versions, etc
the app shows on the device but every time i open it im met with this
just using ar foundation and arcore to build ot android
i had gotten it to work once yesterday and ever since it has failed
build alone succeeds just fine
nvm solved it. for anyone searching this same issue it was on an s22 ultra and i had to disable verify apk/abd over usb. patch/patch and run still doesn't work but build and run from build profiles does work on the device
Fwiw, we use aframe / mindar, and dont have to pay anything. It doesn't work for all projects but it's been phenomenal. Plus being able to fix a clients AR thing in seconds and just asking them to refresh to access it is super impressive.
this basically requires the user-facing camera to have depth so you can simply place it behind the user's head, or you make an invisible mask and then have that "mask" literally mask the object, so you're getting a basic shape that erases objects behind it, this is a more shader-focused approach and may be more difficult to exactly define "correct" here, but it would be a decent idea if you're without any ideas.
I tried a chat GPT approach of adding a shader with depth buffer but it seems not to work
It seems like a solveable issue, I'm not sure how with the details
that's... not what I meant by depth, chatgpt can only help if you already know what you want, because it it doesn't work, you're going to get talked in circles
AR Foundation supports depth in most world-facing cameras, which allow you to occlude objects in your scene with real objects, but I don't think many, if any at all, phones support depth in user-facing cameras
which means my next best guess is to make a game object like the face mask sample scene, but instead of it being visible, the face mask is invisible and it's cutting out the objects behind it, which then allows you to place objects behind the mask in your scene and they will act like the face is occluding them
That's a very plausible approach, have you tried that once?
I'm not new to graphics, but I'm new to Unity, therefore am not sure how to exactly implement it via unity
I forget the exact keywords to accomplish this, I haven't tried it but I am sure it's possible, maybe search for something like stencil masking, or make "cut out shaders" or something, you're looking to find a way to make an object make another object behind it simply not render through it, while also not rendering the closer object at all.
tutorials online may help, sadly I don't have any easy ones on hand
Thank you for your replay, it does sound a good approach but again I'm too newbiew in unity,
I'll be done with the Mobile AR tutorial and then do some basic unity tutorials
Thanks š
Hello, I'm currently using the Unity Learn pathways and I'm on the face filter tracking part of the segment. I tried doing it and it didn't work, so I went back to the first mission which was a simple block floating around. But now it seems like nothing works. Can someone help? The errors relate to xcode expecting angular brackets over quotes, and unity framework. I'm assuming this has to do with xcode not being able to detect or use the unity packages.
"the errors", a mysterious movie about a string not being shared š
Sorry, could you elaborate?
The errors relate to xcode, thwat errors? You need to provide some information for people, otherwise, we could just be doing guess work on all the thousand error codes and information xcode can give you š
First easy steps, clear the build folder in xcode (google for the shortcut, becomes handy) and make a build with clean build, so little arrow next to the build button, clean build... and try again to rebuild. also google for the bitcode error. Might just be the unity tutorial being a bit outdated or not covering it.
Okay! Tysm. Iāll try it.
Can someone help me out with a vivox query?
I am trying to use vivox with Hololens 2 in unity. Will bringing the UI component, VivoxVoiceManager and AudioTapsManager component from the package's main scene be sufficient for it to work with Hololens?
I tried it but the button interactivity is something to work on now.
Meanwhile, what would be the easiest way to make a unity UI button intearctable in Hololens?
I never worked on Hololens through Unity but I guess, it should work out of the box in worldspace when you enabled the correct packages
ok
The login is to convert any UI button into one usable by Hololens. I will be making builds for testing purposes (Not the remote testing thingy).
Do you know how I could convert a UI button into intearctable button?
Thank you so much :3
Hey all,
I know this a sneaky way but
Does anyone have access to dissonance and could lend it to me somehow? As I have explored Odin and Vivox. So far, none of them are working out on Hololens and I hope to god that dissonance helps me out š
Hello, I'd like to create this effect in apple vision.
It's a project to show visual deficiencies, in MR.
The idea is to have a spot that follows the user, and where we would see through the blurred and distorted environment. My problem is that since the scene is in MR and I want to retrieve data from the environment, I can't use a Scene Color node in Shader Graph. Do you have any ideas? Do you think I can use the Polyspatial Environement Radiance node with a distortion node like Twirl to achieve this effect?
Checked out agora?
Should be possible, if you just fake passthrough
You can generate a mesh of the environment, but you can't be sure of recovering the textures and lighting, which is precisely what you need to achieve this kind of effect... What's more, I'm not sure the Apple vision can handle it.
One of the big problems I'm having is that the camera data isn't available, so it would be easier to modify the camera rendering using a shader...
Afaik this is not possible without camera access. I have been trying as well on various headsets and made an okay prototype with the Pico 4 Enterprise
It is through os 2
The node polyspatial blur seems to pick up elements of the environment, light tones etc., but it's very hard to tell as the effect looks very very blurred. I was thinking that by starting from a plane with a shader using this node and some way of transforming the rendering, we could have come up with something.
Ah? I'll take a look
Is it well supported by apple vision?
I don't support piracy, so I'm not going to send you a $100 asset. But it does list hololens on its page, and if you need a trial, ask the creator of the asset.
You should be able to get a refund if it doesn't support it though
For hololens we just created our own though. It's just packing of bytes and adding a small buffer. Works well for us. But off dissonance is likely going to be better at scale.
You could also see if they have a discord to inquire about it
Sure thing. I will see if they have a discord and also I will email them for a trial version
If it works out, I might as well give the 100 dollars.
Anyone in here working with Unity on VisionOS and knows how to implement Shareplay into the exported xcode project?
Hello everyone, I'm new here. I want to ask a question. So I'm working on an AR Project to place posters on the wall but encountering some problems.
When Instantiating the prefabs my poster is sticking out of the wall
I'm using this code :
void PlaceObject(Vector2 touchPosition)
{
if (raycaster.Raycast(touchPosition, hits,
TrackableType.PlaneWithinPolygon))
{
Pose hitPose = hits[0].pose;
Instantiate(placedPrefab, hitPose.position,
hitPose.rotation);
InteractionController.EnableMode("Main");
}
}
I followed a tutorial from this PDF, and now my poster is in the correct position, but the picture doesn't appear.
void PlaceObject(Vector2 touchPosition)
{
ARRaycastManager raycaster = GetComponent<ARRaycastManager>();
if (raycaster.Raycast(touchPosition, hits,
TrackableType.PlaneWithinPolygon))
{
ARRaycastHit hit = hits[0];
Vector3 position = hit.pose.position;
Vector3 normal = -hit.pose.up;
Quaternion rotation = Quaternion.LookRotation
(normal, Vector3.up);
GameObject spawned = Instantiate(placedPrefab,
position, rotation);
FramedPhoto picture =
spawned.GetComponent<FramedPhoto>();
picture.SetImage(imageInfo);
spawned.transform.localScale = new
Vector3(defaultScale, defaultScale, 1.0f);
InteractionController.EnableMode("Main");
}
}
Anyone know why my picture is not appear?
For image tracking only AR, I don't use Unity anymore. But my guess would be either your picture is backwards and that's the back, or you configured your material wrong.
It's common to have a Base > Pivot > Frame structure in your prefab to be able to resize and rotate the model
I'd try using the AR simulator to see in the scene what the rotation is doing, or add a 180 rotation on the child model and blind test it.
I have a structure like this. Do you have any suggestions on how to solve it?
Have you looked at the back of it to see if that's the problem? Id start there to make sure first
Just because there are other possibilities (I.e. Z buffer)
Yes I did, but my poster is not on backwards
I don't have enough information to help you, but maybe someone else will chime in
Anyone know how to get RawImage from this ImageButton object?
anyone thats interested in building tracked devices with the MX Ink, I did an IR coverange map, and tracking tests with the device: https://www.youtube.com/watch?v=OfGSaZKiaCI
We finally have the first trackable object for Meta hardware . . . So now we have to build a new pistol controller for Augmented Defender.
Part 1 focusing on tracking, mounting, and initial testing.
Game:
https://www.meta.com/experiences/augmented-defender/5128569437244342/
Can anyone recommend me a good tutorial for AR Fundamentals? I am planning to make an AR game but I have no idea how that works?
!learn should have one
:teacher: Unity Learn ā
Over 750 hours of free live and on-demand learning content for all levels of experience!
GetComponent<RawImage>?
Hello, is this the right channel to ask for help? It's for my AR Sign Language Research. Thank you š„¹
Do not ask to ask, just get your issue in here and youll see, if someone can help. If its for AR, the AR channel seems to be right š
Okay, thank you. I was just making sure because I can't seem to find the help channel if there is one : D
I just wanted to ask how to put speech to text and text to speech in my AR. I followed a YouTube tutorial first but the code seems not to work.
If it's alright I can send my codes. I just really need someone to guide me because I'm a beginner ಄ā āæā ಄
I am not sure, how AR and speech recognition have to do anything in common. If its not AR related, you mgiht get faster help in any coding channel #š»ācode-beginner #archived-code-general and so on. And if you want help, you should explain, what "seems not to work" mean. Just not working does not help anyone helping you š Write down, what you have tried, what the errors are you get. And your git repo is private i guess, because I get a 404 on it
Yeah speech recognition for mobile is probably all you really need.
For hand gestures though I'd just use OpenCV. Cross platform, works on older devices, and has hand tracking
AR Foundations fairly pointless beyond plane and image tracking. Face etc is all locked to only iOS.
Ah correction they finally support face tracking on android it looks like.
That's good to know
Ooh thanks, this is probably what I need. I'll also try to set my git repo to public this time, sorry I didn't notice š
Hi guys, really new to AR, im having a problem with things being imported from blender, i was testing with this little cube, i exported it as .fbx, and when i try to place it on a surface, suddenly this shows, its just that still image, any idea if its related to how it is being exported from blender or any camera setting i need to know for these? With built in models of AR it works perfectly (built apk)
This doesn't look like an AR scene. Is AR on and set up properly in the scene?
Are you sure you didn't export/import the camera from blender as well? You can change this both in blender and Unity import settings
Yes, this is the built apk, the problem might be the camera export u mentioned, i will try again later, thanks
hello I need help, I'm kinda having a problem in my game rn. the plane detection in AR scene are having a problem detecting plane I dont know ehats the problem
the thing is at first when I only have one cat to use it did a good job detecting plane. but after a while I added another cat,its now having a hard time making a plane in AR scene
Im using Niantic for the AR
Hi, i have made a test project for my Apple Vision Pro and added XR Hands to it. I want to grab a ball by a grab or a fist gesture not by a Pinch which feels not naturally by throwing a ball. I have implement XR Hands 1.5.0 and the grab gesture will be detected (the sample text will show that) but I don't know how to step further, how to grab something with this gesture.
I do not have much experience in Unity so it's hard to step in. Is anybody able to give me a hint how to realize that?
You just have to combine what you want to achieve. If you want to grab something, you got the hand gesture grab and from there, you can do whatever you like. check against the distance to the object, parent it to your hand and rotate to make it work with your palm and so on.
Guys I want but AR/MR device but I don't know which one is better š
Imo
The best hardware for the price is the Quest 3, or if you are on a budget the Quest 3S
If you don't want a Meta device, the Pico 4 Ultra is a great alternative, just with less market share
Thx, but I am too stupid for that
Well, maybe game dev is not for you then. Putting no effort in will outcome in the same amount.
@nimble fossil I don't know how you came up with this, but I've already invested days in this topic. The thing on Unity is that you learn something and two weeks later all is deprecated. I know, I can add an Collision sphere to my hands, on Gesture Event, I can read get the Object which is in collision and add it to my tree to simulate the grab. But I guess there must be a build in solution with visionOs, XR integration Toolkit and XRHands and that's what I'm asking for. Thx
I have seen the "Unity VR/XR Developer: Make Immersive VIRTUAL REALITY Games" course from Udemy but I'ts complete different as in my Sample from visionOS.
You are better off trying out things and testing instead of doing those udemy courses. Especially with a quite new hardware that gets updates on OS side and unity side with each package update. But I am working on the visionos for months now and I can tell you, its not, that everything gets deprecated all the time. If you watch old tutorials, then I agree. But if you stick to the docs, you usually get either quite stable APIs OR your IDE is already giving you results for those issues when something is marked as deprecated.
For your issue, you get from visionos INSIDE unity the tracking of the hands. When checking out the XRHands samples, you also get gesture recognition. So not sure what you are still asking for, as you got all the tools you need. If there is a visionos native support, it would basically do the same. Check your hands palm collider against items in the scene and grab it when hit and detected grab gesture. But as we are working in the Unity environment, you just have to mimic that with your system. There are maybe some grabbable scripts already somewhere, did not investigate everything in the sample packages. Also the unity vr multiplayer template gievs some scripts that MIGHT work with visionos, something to investigate too. But "invested days" is like the minimum you hhave to understand, what you are using anyways. As said, I am working on this for a very long time and had to hack my way around a lot of things because it was not yet supported or only partially and so on. Thats dev life š
Hey, to anybody who's developing for Vision Pro in Unity, is dynamic lighting from the real-life environment supported in some way? Also, I've seen some demos with dynamic shadows on real-life surfaces which was nice. I'm really bummed about the lack of support for VFX, I hope this gets resolved in the future.
For the first real life lighting, you right now get IBL lighting and reflections out of the box, which works quite well already. About shadows, I assume, they might be using some mesh detection and a virtual light to achieve additional shaders.
Hello! Is anyone familiar with AR Foundation Planet Tracking? I'm working on it right now, but I've encountered a problem while trying to detect more than one card planet. The prefabs are not instantiating. I'm following a tutorial where I'm setting the Tracked Image Prefabs to None Game Object and using the Planet Main Mode Script Mode. Does anyone know how to resolve it?
What the heck is planet tracking?
Hi I'm making AR with Vuforia
I made a hand dance but had problems because there was an area of āāthe hand that came out
I've made a simple masking, in the editor it works well but after I installed it on Android the area that should disappear instead appears in black
Is there another way to hide objects that come out of the area I specify?
Hello. Does passthrough in unity editor work for you? I'm trying everything and don't see it in editor but works in build or in HMD during play. Or cant I see it in editor?
What passthrough? Like, what passthrough do you expect in editor to show up?
I'm using Meta quest 2/3 and I want to stream the passthrough what is shown in HMD into editor when is in play mode
And what did you do, to try that? I did not work with the quest a lot, but is there an option to stream or did you just expect it to happen?
Expectation. It would help me with testing. But if this can't be done, I'm ok with that. 
seems I am wrong: https://www.youtube.com/watch?app=desktop&v=sSD798Ov2oY&themeRefresh=1
How to set up Oculus Quest Link for Unity's play mode. Go in and out of play mode for your VR game without needing to build and deploy to your device!
I'm taking a 1-video break form shader tutorials this week to briefly talk about Quest Link in Unity. If you are doing Quest development in Unity this is huge to iteration. Especially if you a...
Yea I know this, but this isn't passthrough :/
any free alternative of WebXR plugin. So that AR can be used in IOS web?
Did you enable it in the desktop app? I think it's still in the beta section iirc
We use aframe, mindar, etc. You don't need unity for that
ok will check thanks for pointing out ā¤ļø
will it allow ground placement without any image track or anything?
like AR foundation?
Yes - there are examples with it, but also boiler plates like
https://github.com/chenzlabs/aframe-ar-plane
There are some webar plugins for unity, we bought em all, but I haven't seen one work overly well
yeah it sucks i lost a big project cuz unity does not support web xr that well
At Unite in Barcelona Meta showed the Quest 3 build profile. Have we heard any update on release for that?
Hi,
I'm currently working on a project in Unity where I'm using Kinect to animate a clothing model in real time. I've successfully imported my clothing model from Blender, and in Blender, the armature correctly moves the clothing, including the sleeves, in Pose Mode. However, after importing the FBX file into Unity, I noticed that the sleeves of the clothing are not moving when the arms move. The rest of the body seems to be moving well, but the sleeves are static. Can anyone help?
So if you move the bones of the arms, do they move the shirt?
Yes the sleeves should move along with the arms, but now only the shirt moves the sleeves remains static
What meshrenderer are you using? Skinned or default?
Default
For a mesh to move with bones you need a skinned mesh renderer that's set up by importing a model with a skeleton
The regular renderer component can only move the position of the object and render it without bending anything
Hey everyone š . Does anyone know if there is an AR tool/framework for Unity that supports foot detection (or body detection in general) that also works for Android (i know about AR foundation body tracker, but it only works on iOS, i'm looking for one that works on Android, too, if it's possible).
I see i'll try to fix that, thanks bro
Hello everybody, anyone have an idea how to make the background of a video player transparent? I have multiple video players in my AR project and the videos they render have a transparent background, but the background of the player itself is black which makes it seem like the video has a black background. Instead, what I want to have is a completely transparent background so that it looks somewhat like a hologram. How can I achieve this?
thank you, i already have a transparent video as i mentioned, so i already went through this documentation, what i am trying to do is different but i think i might have figured out what the problem is and how to fix it. thank you.
ah okay, why would you let anyone know then, what your fix is š
How do I set up Multiplayer XR in Unity?
Meta Quest 3's -
It requires an OVRCameraRig to be in the scene to even start the application, right?
But if the player prefab is the OVRCameraRig - how do I start the application and connect players to eachother?
Do I have to have an 'initialization' scene which has the OVRCameraRig that then transitions to the real 'game scene' that instantiates the player prefabs? I'm not sure how the specifics of that work I can't find a guide online either - its for XR specifically.
It does not really matter if its XR or not. Its basic multiplayer tutorial that will spawn players on their positions. the local player will have the camera rig and all others are just representations of remote users visible for the local user. You really should look into the vr multiplayer template provided by unity.
Okay so let me explain maybe someone would need this, I have a canvas that has a scroll view that has multiple video players. At first, my videos did not have a transparent background because they did not support the alpha channel, so I used Shotcut, a video editing software, to remove the background and export the video with the WebM extension as in the Unity documentation it is mentioned that this format supports alpha channel. So that was the first step towards achieving a transparent background. Now the issue that remains is that the background of all the parents of the video player, including the scroll view and the canvas is not transparent. So, the solution I came up with is to change the sprites used by the scroll view viewport and the canvas to something that has just an outline, and any other components that have a background color bring the alpha down to 0. I have some issues when changing the sprites, but this is the cleanest solution I could think of. It is a very specific case, but I need this transparency because I need the person in the video to be as if they are sitting or standing in the real world. I hope this is clear enough.
Hello guys. So I'm sort of new to AR but I've worked a little on manomotion for a bit.
I am building a lab simulator showcase for school using hand tracking on android.
Would mediapipe be a good choice for hand tracking on android or should I use any other library?
I dont want to use manomotion again as its a bit unstable
We use opencv with media pipe. I had to do a lot of work to make it performant but happy enough with it.
It's good for up to 6feet. From there I move it to skeletal pose tracking and that's testing well for up to 40 feet or so.
I see. Do you have any boilerplate for the config?
Or a repo maybe
No, I gutted and rewrote a chunk of it, and redid the Unity samples. There was some questionable code and bad math on how the Enox integration was done.
You'll see once you get into it
I don't really fault them. It's converted from camera to screen to world coordinates. And you can tell they are more academic and less of an experienced Unity dev. It's enough to get the basics working though
I see. Also how exactly does the export to android work?
I've seen people use external python socket connections onto unity for hand tracking
Is it easy to export on android?
Pardon me if I sound a bit uninformed, im a bit new to AR
It depends I think on how you are accessing media pipe. In our case we just modified the OpenCV and it's all built in. No sockets, no other things. The model and everything is built in.
The existing OpenCV from Enox has an example with all of that working, you just do a normal android build
This seems to be paid
Is there an open source alternative?
Im able to run base mediapipe but it just seems to be a 2D video relay. The 3D stuff seems to be based around OpenCV from Enox which seems to be paid
https://youtu.be/RQ-2JWzNc6k?si=bWr4JYCKZ8_00frv
this guy seems to be running hand tracking with opencv in python connected to unity via UDP
would it be possible to export something like this to android?
Visit http://brilliant.org/Murtaza/ to get started learning STEM for free, and the first 200 people will get 20% off their annual premium subscription.
In this video, we will learn how to track hands in a 3D Environment. We will write the Hand tracking code using python OpenCV and CVZone. From there we will transfer the data to our unity 3D Env...
Anyone know if passthrough works for PC VR
Does anyone have advice for implementing large animations in AR?
I have this animation that I want to place in an AR environment but because I want the video background to be transparent (and I don't know of a way to implement a video with transparent background), I'm having to load thousands of pngs with transparent backgrounds, each representing a frame of the animation. Right now, I'm trying to break the animation down into smaller chunks and dynamically load the pngs as needed, but my app still keeps crashing from memory overload.
I would appreciate any adviceš
On Quest you can do it if you enable developer options in Link and passthrough there
VP9 has transparancy support, so that might help. Otherwise adding a shader which discards green pixels works as well if you can green screen it on a regular MP4.
I would definitely go the video route instead of so many images
Yeah ok. Thank you
Hi, I'm stuck with a problem I never seen before on Unity. Maybe you have some suggestion to give me ... I found no solutions for now ...
I'm working on a Mixed reality app for Meta Quest 3 using Meta XR SDK.
I started the project on Built-In pipeline, but after few days I decide to convert it to URP so I do:
- install URP package
- create URP assets
- bind it in quality settings (I delete all quality preset exept medium)
And now ... when I start my app, I have no render except canvas ... all mesh renders not render at all...
Do you ever eard about a problem like this ? In 5 years of developping on Unity, I never see a problem like that š¤Æ
Did you upgrade your materials?
Anything in logcat?
How does it show in editor?
If you have the pose positions from the 2D video relay and their respective camera coordinates, you don't really need OpenCV anyway. It doesn't really do a very good job on world position translation. We figured that out the hard way (depth Z based on scale of the subject). It's not the best AR solution (as it doesn't really support culling etc without a lot of heavy processing) but usable if you are in a bind.
on editor (via Airlink), all work well ... the problem occur only on build š
Then I think it's best to update Unity if possible, maybe switch graphics API, and check logcat
I will try that, thanks you for your advices
Hi guys, has anyone heard of merge cube (https://mergeedu.com/cube) , i want to replicate the implementation of such a cube using Unity using arcore and arkit(without depth). How shoudl i start to approach this? Thanks in advance!
I am currently getting this weird error in Unity, when building for AVP: Thread 1: Swift runtime failure: Unexpectedly found nil while unwrapping an Optional value along this thread part and also this error in the logs:
/Users/bokken/build/output/unity/quantum/Packages/com.unity.polyspatial.visionos/Source~/PolySpatialRealityKit/PolySpatialRealityKit/MessageHandlers/PolySpatialImplVideoPlayer.swiftCan't show file for stack frame : <DBGLLDBStackFrame: 0x34b060bb0> - stackNumber:1 - name:PolySpatialRealityKit.resetVideoPlayers(_:). The file path does not exist on the file system: /Users/bokken/build/output/unity/quantum/Packages/com.unity.polyspatia
Looks like there is a bug in Unity/visionOS. Whenever I got the visionos video component running, take off the headset and set it back on, it breaks the app. Only happening with an active video player
yeah i got it down to a basic level rn but I feel like im facing the same world position translation issues as you mention
its just the 3D nodes and the position of the hand on the screen not matching
unsure how to fix that
Yeah, we went through that too.
The problem's a bit complex, the first problem is camera to screen. .. you usually have a 640x480 (or something other) camera feed, but the phones screen resolution is different (1920x1080 for example). You need to calculate that first to get the camera-to-screen ratio for offsetting the position. If using the OpenCV package, discard everything they are doing with bone tracking - it all doesn't work - and get the numbers from the raw camera feed. I used some small cubes to track key bones, until I had it tracking perfectly on the screen.
Now that you have the positions tracing properly, you have the second bigger problem, and need to determine the depth. In our case, I only had 3 days to work on the project, so I guessed on the Z depth based on the length of a bone versus what the average length should be. It gives me a rough percentage of scale, which can be used for a Z depth that was accurate enough. I tricked this by adjusting the objects scale instead of the actual Z position - but that's because the scene was also setup orthographic which would discard Z depth anyway.
Finally it all is very jittery for Z depth, scaling, etc. So I added a 6-10 frame "average" where it would have a target destination, and target scale, and smooth towards it. That seemed to really round out the rough parts. And by then we were done day 3 and go-live went well. Still, very easy to understimate the effort. And it wouldn't have worked if I didn't tear it into bits and work through it ... It's not a trivial task
But it did turn out great and was very impactful at the end
Damn I see. That seems like some heavy lifting.
Ill try getting this done and get back. Also how would you attach the solution to an AR camera rather than the current canvas arrangement, if I'm planning a project involving plane detection
By default its based around a canvas with some annotation layers and a raw image with a camera feed
You can use the arfoundation camera and pass it into opencv or media pipe if you want, I know there is an opencv example that does that. I did do that once a long time ago, and ended up downsampling the camera before submitting it to opencv
I see. I'll test that out too. Thanks a lot man
Hi, I am getting some compilation errors in unity scripting, can anybody help me? I am a beginner
How is this related to augmented reality? And do not ask to ask. Put your issue in the correct channel and just describe with your code, whats the issue. As you are asking about scripting and you say, you are beginner, you might find better help here #š»ācode-beginner
Oh nevermind, you just crossposted... #854851968446365696 do not!
Hi, I am facing these errors while running an AR Project. Also, I tried to build it in build setting and then tried on my android mobile device. However, the camera didn't work and the screen of my mobile was all black with one button option for which I created UI
Like this , my mobile screen went black
Please help me with build and settings where I went wrong?
Anyone have experience using 3D height maps for large scale worlds (Iām talking the entire country or Ireland at 10m per pixel)
Im making an AR app which overlays tapography over the real world
Are the warnings correct that you need to enable a provider in project settings?
I tried to make AR project in Universal 3D Core template
Everytime I try in any updated unity version, there are always different errors and some of the new errors comes out
I fixed this but now the problem is when the sleeves moves, the whole mesh deformed, any idea what causes this issue?
you will need to make sure you are installing the package for your version of unity, and then enabling the providers so you can run it on, for example, android or ARKit, or enable it for XR Simulation if you want to test in editor, which unity has samples for.
Can you show or explain what you mean by whole mesh deforms? Is it a problem with the model itself?
When I run it without the arms armature, the garment can follow the userās body position, but the sleeves remain static. When I run it with the arms armature, the garment deformation looks like a butterfly. I can see that the arms are moving, but the garment behaves like itās scrapped or mangled.
perhaps this means you need to fix the model?
Do you have any specific question? A general question, if anyone has any experience usually does not lead to any answers at all.
Hello, trying to register AR vertical planes (walls) to hide stuff behind with opaque planes. However, I'm noticing that plainly speaking the vertical recognition is not the best when I test it. What settings should I tinker around with, and are there any additional tools I should use as well? Currently building to android and using the ARCore package. Thanks in advance.
So your platform would be android smartphones? I guess it highly depends on the smartphone itself, how well the sensors and camera are and therefore how well the calculations can be done or the image quality is enough to get wall detected.
Hi everyone! š
I'm working on a Unity WebGL project and need to implement functionality similar to Unity's Input.compass.trueHeading API. Specifically, I want to retrieve the device's compass heading on mobile platforms (both Android and iOS) when running a WebGL build.
Iām looking for a custom .jslib plugin that:
Retrieves the device's compass heading using DeviceOrientation or related APIs.
Ensures compatibility with both Android and iOS browsers.
Handles iOS-specific permission requests for accessing orientation data.
If anyone has experience creating such a plugin or knows the best approach to achieve this, Iād greatly appreciate your guidance or examples! š
Thanks in advance for your help! š
I am using this functionality in webgl ar
does unity 6 even work yet?
Works great for me :P
Hello, does anyone have experience with similar things??
It's a lot of work to do it that way, not impossible, ive had to do a few custom plugins for things related to our WebXR. But because it's webgl, you can pass some of that information in from Javascript to the application.
https://stackoverflow.com/questions/52594247/how-do-i-return-a-compass-heading-in-javascript
https://docs.unity3d.com/2017.3/Documentation/Manual/webgl-interactingwithbrowserscripting.html
How to fix AR Foundation black screen when built and run.?
hello, can ar foundation used for location-based ar?
Thanks for your help!!šš
Using URP with AR??
hi guys is this note still applicable today?
and would you recommend ar foundation 4 or 5? it says 5 is still in beta
Check Unity docs. AR Foundation 5 and 6 are both out of beta.
Use the latest version that is compatible with your Unity version. For Unity 6 that is ARF 6 (or 6.1 preview if you need any of those features, it does have better render graph support)
And yes BiRP and URP both are supported. Since Unity will phase out BiRP support, I recommend to start using URP
You can extend it with Lightship VPS afaik
Does anyone know of a good plugin or something for recording a rendertexture or even the camera directly to file?
Im currently doing it through code taking the frames every fixed update (changed to 0.04 for 25fps) putting them in a buffer and then sending them all to ffmpeg to build a mp4.
Problem I'm getting is that sometimes what's being displayed lags a little bit, but my capture is still constant, so the content in the video then speeds up after its stitched into a video.
I'm trying to work out the most efficient way to do this.
My project is an augmented "mirror" that has a camera feed in and a character that walks on and the user interacts with.
Guys has anyone been able to make a unity instant android app with ARF?
Anyone knows of a particlesystem custom component that converts to a polyspatial realitykit particlesystem?
Oh I guess repliacte properties should be working better now
hi y'all, a quick question on AR and programming. I am trying to make an AR app on the Quest 3 to display a wireframe cube around a person. The coordinates of the person are obtained with an external system (optitrack), and all the app needs to do really is to make this cube. Ideally it would be great to have also some debug information displayed as a HUD, with distance, range etc. But so far I am unable to draw a wireframe cube (I can draw a cube with material), and also I can't draw text on a HUD (I have tried with a Textmeshpro object but it doesn't display anything (not even on the dev window))
I apologize for my cluelessness on this, I am working in transportation. This is for a transportation safety project
(in passthrough)
hi anyone facing login issue on vuforia website?? I recently created a new account, the support link is also not working anyone got any solution for this?
Thanks to whoever is responsible for updating the Unity package for ARCore support to take advantage of the Vulkan support, it's really helpful š
hello does anyone here has used the ARCore XR before or even now? I kinda need a help. Im looking for a way to make the cat wander around the mesh its kinda hard to find a youtube tutorial in how to make it. does anyone can help me out on how to do those?
this is a clip from the game we are developing
Im looking for a code or a way to make the car wander/walk around the navmesh
anyone here use vuforia i want to create a image target with buttons in where when i hover the camera to image target it will show my scene 1 animation then if i click next it just show only page 2 animations and so on
hello can someone help me with my project for developing simple project using vuforia for hololens 2?
im using vuforia i already figured out how to play specific sound when image is track but my problem now in my prefabs there multiple animation controller cause when i play my image target 1 it show scene 1 prefab animation but when i show now my image target 2 the animation in scene 2 is already finish i thought i could use how the sound works but its look like isnt is there way to do this?
can anyone help me my image target child wont show up, idk why
the image target itself is tracked but the child not show up
Just wondering if some of yall have used the Meta XR-all in on with the build blocs. Also I wanna know what do you need to add to make a button fonctionnal on a UI. I try to find some information about it online but wasn't able to find any that is related with the Meta xr-all in one
Anyone having issues with referenceimagelibrires being null when trying to access the texture of the referenceimage? It is read write enabled and keep at runtime is true.
Just drag and drop and copy and paste from their sample scĆØnes :P
Gotta correct myself. The referenceImage.texture is empty, not the library itself.
did you load with asset bundles or addressables? Or some other method?
Just straight from the project file structure. I checked the docs, also tried to create a runtime mutable library, but still the texture is always empty or has an id of 0s. The image detection is still working as always. I checked the docs of apple about it and they do not even cover a texture there, so maybe it is just not implemented yet?
Thanks for reaching out btw!
does anyone know any web ar platforms?
You should be more specific about what you want to do
Like, what do you want to do with it, as there are so many to chose from
Hm this is strange, I don't think there should be a problem if you're loading a pre-prepared image library, are you doing this on a platform or in editor? What version of ARF are you using as well?
8th wall, snap studio, and Wonderland all sound quite good depending on the needs
I need white label
Packages are all latest versions, but I also had this issue before with previous versions. ARF is on 6.0.3. The issue is only happening on visionOS runtime and I guess, there seems to be some hickup in the conversion, because also the order of the images is changing when building to the device. So I have 5 markers called Marker1, Marker2 and so on and they are fine in editor, but as soon as I build, its 2, 1, 4, 3, 5, for no reason at all š
a log created on runtime gives me this order for example:
[{"referenceImageName":"Marker2_FF9EDA81-4E28-7B4D-8F13-BA20AD3DCE4E","trackedImageID":"","trackedAnchorID":""},
{"referenceImageName":"Marker1_93C4CA59-C7FE-3B49-B3F4-2395B537C26C","trackedImageID":"","trackedAnchorID":""},
{"referenceImageName":"Marker4_2C4D5ED7-DAC3-4744-A156-9940D14BAD9C","trackedImageID":"","trackedAnchorID":""},
{"referenceImageName":"Marker3_9B713856-55E4-D440-B471-BA02912AE094","trackedImageID":"","trackedAnchorID":"474664A3B7C8F747-F2F9ECA3BFED859E"},{"referenceImageName":"Marker5_3BB965F8-FAAB-0541-8DAC-232CE07F550F","trackedImageID":"","trackedAnchorID":""}]}
You can see it starting with marker 2 and so on. But on editor, its logged just fine (screenshot)
I like self-hosted & I'm not an overly big fan of arfoundation. I usually use aframe or mindar. But it depends a lot on your context.
I want animation and interaction
I assumed you would, it doesn't change my response
I need ground tracking with custom loading screen and everything customisable its fine if its paid too around 100 usd max ig
I'm confused. Aframe supports ground tracking, it's web so you can add loading screens wherever you want. And it's open source and as customizable as it gets. I'm not sure what to clarify for you, but you may want to search on the web on your own on that case or use chatgpt.
Anyway, hope it helps, that's the best I can suggest with the requirements - I wish you luck in your journey
Hi, does anybody who is using Unity OpenXR Meta and AR Foundation knows a way to filter only the Planes in the Space (Room) you are currently in, for some reason all the Spaces you have setup are loaded (a bit annoying). We want to show a preview of the current room you are in
hello i have question related to vuforia image target, is hololens 2 have maximum image size for image target? because when i use my original size image target the child of the target wont show up, but when i change the image target size from xml file, the child now show up
note : when using the original size image is still tracked but the child wont show up
Are you talking about within a house and multiple rooms will be loaded because they are all locationally near each other? Or do you mean something else?
Hm, you may need to go ahead and file a bug if nothing else comes up regarding your searches online about vision OS and this behavior, or if nobody else has any experience with this. Either that, or you can check the unity discussions (forums) and try a post there, more of the devs are possibly available there.
Alternatively, check if your build has any logs saying the images are low quality images, it's possible they may be removed depending on the platform if they are "low quality" (determined by platform), but I'm not sure about which platforms do what.
Thanks for trying to help. Thanks to the industry plan, I am in touch with your colleagues now and we are trying to evaluate, what the issue might be. Ill let the channel know, if its something, that needs a workaround or just an issue on my side š
Thank you! Good luck and hope that it comes to a fix or a solution
I think that's the first time I've ever seen someone say thanks for the industry plan
Yeah, thats the thing, I dont know what exactly is the internal behaviour, does it show every space setup or just the rooms that are close? is there a way to exclude planes that are not part of the current space I am in?
that one I'm not quite sure on, sorry, I would assume that quest is loading nearby room spaces because it considers them part of a "house" or something together, but I'm not sure how the spaces are set up in that manner to reproduce it
Stick to the good sides I guess š
hey! my partner and i have been developing an ar poker app and all was going well until tonight. i've gone through and checked my entire ar rig as well as our render pipeline, but cant seem to find anything out of the ordinary? it only happens on mobile as well, in editor the game runs just as expected
the devices camera view isnt even being displayed in the scene. that happened to us before, but setting our render pipeline back to URP Performant fixed that
Did you log anything? Looks like the camera is not clearing up. Check your render pipeline for the needed background feature added. Not sure what unity version and packages you are using
Yes it sounds like you need to enable background rendering feature in your scriptable rendering feature asset by adding it as a component, you can't just swap render pipelines without any changes unfortunately, after that if that didn't fix it, check your camera rig and make sure it's set up to clear the background with a color or something, but that's only after we confirm you have the ability to render the background at all
Ping, sorry forgot
Hey guys, has anyone successfully built a standalone gaussian splat player for quest 3?
@nimble fossil @true stag i ended up finding a solution, saw that color space was set to linear instead of gamma, so i swapped it and set the ARCameraManager's render mode to after opaques
for reference, was unity 2022.3.47f1 with the mobile ar template (ARFoundation & ARCore) running on URP
I'm currently having an issue finding tracking in AR. Anything suggestion is appreciated!
Not myself, but it's definitely possible
There are many GitHub repos, like
https://github.com/aras-p/UnityGaussianSplatting that report to work
@loud kelp
What issues?
Just can't find mobile AR tracking for the body
For iOS?
Android doesn't have this natively afaik @hearty widget
Android but damn
š«” I'll look into it
Is it possible to make a mobile AR VR setup? I am able to create a game that uses just AR using Uinty XR plugin. I have also separately set up google cardboard VR. How can I make an app that uses both? I want my AR game to be seen like the way you see two cameras on youtube videos
Not sure what you mean "two cameras on youtube videos" but basically you can setup AR and VR in two scenes and show your content in both, then just switch between the scenes depending on the platform for example
This is what I'm lowkey trying to do ā¬ļø
thats just basic google cardboard, right? https://developers.google.com/cardboard/develop/unity/quickstart
Hello, I am new to unity and I would like to ask you guys about something. I want to create an augmented reality scene where I can detect a door or a window and replace it with a new one from an image from a database. How complicated would it be?
Very... if you have no knowledge at all
Thanks. Okay, well I am an experienced web developer and a cs graduate. What aspects would be the most challenging? Like Identification of the door position?
I guess the most time would be to get the knowledge, not the technical difficulty, as, depending on the platform you gonna go for, a lot is already covered. But yes, tracking is always a big topic, when you want it to be accurate, cover different devices and what not
We use opencv. The Unity project for it is buggy on the rendering though, you'll need to put some time into correcting it.
You'll need to also use OpenCV. Arfoundation doesn't do object tracking on android, and has no built in way of doing object clarification
With arfoundation it wouldn't know a window and door was there. And even if it did, it couldnt tell the difference.
But the good news is that's what opencv is exactly built to do
I'd argue though you maybe don't need to use Unity at all, and use a web based opencv approach.
Thank you very much for you helpful insight. I guess I need unity to place the door on the wall anyway no? Because the origin idea is to replace the door after I identify its location.
Not necessarily, id probably use aframe & glb files.
You can use Unity if you want though, just need to use it via opencv
Even if i am using the camera because I want augmented reality
Yes, aframe let's you do webar in your browser
Can I import in a flutter app or in a react native app?
It even has better cross platform compatibility than Unity for AR
In Unity? You add aframe the same as any Javascript library. Unity doesn't allow for webar at all.
No I meant can I import the opencv code to use it in a flutter/rn project
There are some workarounds that let you add afrsme to your unitywebgl project. But at that point, you don't really need Unity at all.
I'm sure you can, it's just a library
OpenCV is just open computer vision. It let's you extract information from camera feeds, videos or images.
Such as body posing, hand posing, what an object is, the shape of doorways or windows etc. It's used by a myriad of platforms including robotics
Yeah I know about OpenCV, I used it before in the university.
Thank you very much for your help
Cool, that'll help
Unitys default ar is arfoundation. Which is pretty much just here is a floor, here is a wall, here is an image. With the image one being a bit buggy.
Beyond that the features start working on only one platform, etc. So it becomes less usable for consumers.
Alright, understood.
Ill look into it!
The reason I asked this question here is I have a potential client that wants to create a mobile app with AR. What he wants is to replace a door with another one, and to paint walls with different colors. I told him I had to dig into that and see how complicated those tasks are because his budget is quite low.
So I now if he only wants to implement painting walls or floors would I better go with Unity or stick to OpenCV?
https://docs.unity3d.com/Packages/com.unity.xr.arfoundation@6.0/manual/index.html
Near the bottom of the page, it'll show you what arfoundation provides. None of it works on the web.
You can layer opencv on top of arfounfation but it's tricky
You'll need to go opencv no matter what.
It's just a question of with unity as your engine or aframe as your engine
- Unity/aframe can do floor and wall tracking.
- OpenCV will let you find the shape of objects and identifying objects in the camera.
Keep in mind though, if it's a non-game application of AR, unity will require you to use Pro (no personal edition allowed). And if that client has over 1m in revenue, you need to buy Industry. The best documented OpenCV for unity is also a paid asset, but has provided free updates for ~6 years, so worth it (especially if you can get it on sale)
I saw more indies getting their licensed disabled on bluesky over this recently, so figured it's worth cautioning.
Thank you very much for your help and time.
hey guys, im making an ar app which displays terrain over the real world. i have tiled my map, but need to load the user on a specific part of the tiled terrain to be accurate with their real world loaction. However, im unsure how to grid my unity object with real world coordinates. anyone have prior experience with this that could help? thanks.
Not quite sure, what you mean with real world coordinates? Do you have specific tiles for specific location coordinates and you want the user to see this tile and the surroundings accurately? And how much of the world do you cover? Do you have access to GPS on your target device?
hey guys, I'm working on a project as a complete beginner. I'm trying to build a webAR service for which I'm facing a lot of issues in XR plugins and to integrate AR part in the unity
can someone personally connect over and guide me with the same, it would be really helpful for me. thanks in advance!
I don't think anyone has the time to personally guide you through your problems, but if you post more details you'll get plenty of eyes on it and hopefully some good advice and pointers on how to proceed.
Is it the basic setup you're struggling with or specific issues? If it's the former, it might be worth asking broadly if people have tutorials they find useful and could recommend.
Unity Learn has a pretty up-to-date course you could try following.
https://learn.unity.com/project/get-started-with-ar?uv=2021.3&courseId=6346b7e5edbc2a7caec0f8fc
In these initial tutorials, you will get ready for AR development. First, you will learn about the field of AR and hear from established creators in the AR industry. Next, you will download all required software to get started and configure a Unity project for AR development. Finally, you will deploy a test project to your mobile device to make ...
Thank you for taking the time to address my queries.
Iāve tried various methods to integrate XR plugins, but some of them, such as WebXR Interactions and WebXR Export, resulted in numerous log errors.
I followed the instructions from this video for the above methods: YouTube Video.
Iāve already created an environment (a museum structure) that I need to augment on my website, but Iām currently stuck with the plugin setup. Iām unsure how to proceed without relying on paid services.
ohh thanks for this
I'll look into it, btw it's not an App based AR?
Oh, WebXR, sorry I completely missed that. I'm not aware of any first party WebXR solution. As far as I know, Unity doesn't support AR for Web at this time (Although I might be wrong).
ohh
yes I've also came across the same understanding
there's this third party AR support like zapper, 8th wall but they are paid services
This article is a year old but might be of interest to you ?
https://medium.com/@depanther/webxr-export-and-unity-xr-sdk-how-and-why-d024108f3d76
How support for Unity XR SDK was added to WebXR Export
ohh thanks I'll look into it
Can I host my project on 8th wall's basic plan and later on can upscale it to the pro plan if needed
Yeah so Iāve made a 3D terrain representation of irelands coast. And I want the terrain to move and load for the user according to their location on the phone. A whole load of issues with map projections etc, but I want to essentially algign the unity coordinate system with real world gps coordinates. Iām calling the gps from unity
So depending on the users real life location, the terrain loads correctly to that area and is accurate
Are you wanting to provide the ability for people to zoom out and see the terrain around them and the like?
We use aframe, three.js and mindar.
No sense paying 8thwall etc if you can avoid it. Unity pretends webxr doesnt exist - and you dont have to have their runtime loading etc anyway this way.
It's going to depend on how you are doing your terrain. Unity uses meters, but whatever you are using for terrain generation should accept GPS coordinates or you can convert them if you know your terrain positioning...
Just keep in mind the GPS can be wildly inaccurate for AR overlays (limitation of GPS) and some android phones return very different base values for their altitude.
There is a (rough) AR GPS package for sale in the store. But at least it's an attempt to tackle the feature if you need a hand up. I think mapbox may have a sample as well.
Also especially on the coast your gps will just be worse. And you are relying for it on location AND rotation. You mind need to rethink your concept tbh. Maybe rely on geo anchors from ar core or something on specific locations but still this relies partially on gps and the camera image.
Thats a dream but technically close to impossible just with gps only
Iām just focusing on coastal waters so gps should be fine. But will look into it
no, just to display a 3D projection of the sea floor at their location
Hey all. I would like to know how to deselect a XR Grab Interactable via code. I am using AR Foundation where these interactables are spawned inside the ObjectSpawner. Now I would like to iterate ower these and deselect them so they're not highlighted any more. But I can't find a method for that.
found the solution: public void DeselectAll()
{
XRGrabInteractable[] interactables = objectSpawner.GetComponentsInChildren<XRGrabInteractable>();
foreach (var interactable in interactables)
{
interactable.focusMode = InteractableFocusMode.None;
}
}
Hi. I;m following a MR course for Unity and Quest. I've come to a part in the tutorial where its saying to add a room model block. Yet in my version of Unity 2022.3.511/1 it's not there. Is it called something else now?
Hi, is there a proper way to get phone GPS location using the new Input System? Because right now whenever I build the app using both input systems I need to press ignore button to acknowledge something might break.
But I couldn't find replacement anywhere.
Also no effect mesh?@amber sail
Should there be one?
Ah the resolution was too low on mobile
It's the scene mesh now!
It changes often
Any idea about the room
Model block?
Scene mesh IS the room mesh (;
Unless if I'm mistaken or if it's renamed again š
Hey, I have a project where I want to track the roof of a car (while you are inside) and cover the roof window with a plane/portal (portal from the samples). I use the apple vision pro and the unity presets for that. Since I am new to unity, C# and Apple Vision Pro I can't get it to work to spawn a plane or any object on top of a tracked surface. Has anyone already done something like that and could help me with it?
The easiest way is to have the user do mapping I think. Havent messed too much with tracked surfaces
What do you mean by that? Sorry really new to all this š I used the mesh example and it mapped all walls etc. With a mesh so it can track the surface. But how can make the AVP spawn a plane or something like that on that surface?
Ahhhh it got the mesh of course!
You could iterate over all triangles, then check if the surfase is both above the user and is (mostly) horizontal.
Those traingles are the edges of where you can place a virtual window. 'Simply' place something within those bounds!
Okay that's a good starting point I will give that a try as soon as possible thank you!
Good luck!
where can i find a help channel? im starting ar game for uni proj. needz some help thx
Just put your issue or question here
oh thanks. right first thing is vuforia. since my phone samsung galaxy a15 doesnt support ar core, i had to switch to vuforia. but vuforia doesnt appear in the XR plugin management system but it clearly shows that i have already installed it in package manager, via git hub url. any help? thanks
Hello š i'm trying to get started in augmented reality development, do you have any tutorials or project for beginners? Thanks you š
There are tons of templates and some unity learn videos. Id say, play around with the templates and read about the scripts they are using in the docs to understand, whats going on. Depending on the platform, also that specific docs help to understand how for example ARCore and ARKit are working. Also what AR are you going for. web ar, which is a different story making you depend on other frameworks.
Hi everyone, need some general advice, cause I'm kinda lost by now 
I'm working on a small AR Unity project and have been exploring how to implement Shared Anchors in AR. Iām aware that the ARFoundation team has plans to add Shared Anchors on their roadmap, but the feature isnāt available yet. Does anyone know how long such features typically take to roll out based on Unityās development cycle?
In the meantime, Iāve come across recommendations to use the Meta SDK, especially since it supports Shared Anchors on Quest devices. However, Iāve noticed a lot of negative reviews (stability, documentation out of date etc) about the Meta SDK in the asset store. Is the Meta SDK really as bad, or is it manageable for a project that primarily targets Meta Quest?
I also heard about building your own solution to synchronize anchors using Photon/Cloud Storage, but Iām not an advanced developer. How realistic is it for someone with moderate experience to implement such a system? Are there any resources, tutorials, or best practices you'd recommend for syncing anchor data across devices?
Any advice greatly appreciated! ^^
There are some systems that do have Shared AR, Like Lightship from Niantic
For MR, Meta SDK has come a long way and building blocks are very nice to work with. Just a bit tedious sometimes
Meta also released a shared space scan that creates a scan of environments between different devices if you wanna google for it. About the timespan of ARFoundation picking up, the problem is also, that things like meta and probably apple will start to implement their own systems to share it (like visionis groupactivities) which could cause the shared anchors to be obsolete. But thats just assumptions. Fast changing topic which I am part of right now myself, trying to figure out the best practices for bigger events and shared stuff.
For mobile AR, on iOS you can use worldmaps, otherwise I think Google's Cloud Anchors are the most robust and widely used, but they're hosted on Google's servers with a max lifespan of 365 days which is a bit of a shame (Whereas Apple's worldmaps you can serialize and host / transfer yourself between clients).
Oh, I think I kinda buried a key detail ā my app is basically Quest only, and the anchors just need to persist within a single session so I hope that makes things easier. š Thanks for the responses so far! Iāll give the building blocks another shot.
When I checked them out yesterday, it felt a bit annoying that they still seem to use the old OVRRig instead of the new OpenXRRig that Meta highly recommends. I couldnāt find any samples using the new rig ā did I miss something? Or is it quite easy to just replace the Rig while still still using the building blocks?
Then you are stuck with the Meta SDK I think
Maybe XRIT has a colocation project on git as well, but not too sure on that
It can just be that Meta hasnt updated some blocks... Common issue sadly
Did you Check on the new unity 6 vr multiplayer samples? Not sure they make use of it
Hello. I have a major problem after upgrading my AR project from Unity 2023.2.2 to Unity 6. I described the problem in the forums in detail. I would appreciate if anyone here can share their experience in the official forums so everyone can find the answer there in the future.
https://discussions.unity.com/t/broken-ar-plane-detection-behaviour-after-upgrading-from-unity-2023-2-2-ar-foundation-5-1-1-to-unity-6-ar-foundation-6-0-3/1580758
So I'm trying to figure something out...
I created a regular app in Unity 2022 for the phone where the buttons reacted when pressed on. I just used "OnMouseDown()" for that gameobject and it worked fine. It worked for both testing in Unity and on the phone.
I'm right now creating an AR app using Unity 6 using AR Foundation [XR Origin (Mobile AR)] that comes with Unity and I'm trying to make a gameobject do something when clicked on just like I did with the other app and it's just not registering. Anyone have any idea what I may be doing wrong or if there's another way I should go about it?
Experimenting around and reading the error logs, I think it has something to do with the Input System because now there's Old and New. Going to have to figure that out.
Id suggest, converting it to the new input system anyways.
How can I make an app like this using Unity? Is there any unity SDK available for this? its scanner app to generate 3d models for rooms.
ARFoundation has a lot of managers to detect planes and stuff. You gotta read the tutorials and example templates in unity and try it out, asy ou are asking for a whole app. Thats too big of a scope to answer, so dig into it and come back with specific questions about what you are missing
I have tried plane detection but it's not like that. this app generate mesh as we scan the room
its a worldmesh generated, thats an ARKit, not sure about ARcore
@nimble fossil find and install ARkit SDK from package manager but can not find any samples about ARkit for unity. do you know any resources for this?
I made an AR app and want it accessible on a browser. I tried uploading the webgl file to Unity Play, Itch and also uploaded the the index and files to page on my website. They all load the Unity loading screen along with the Unity splash image, but then all I get is a black screen on each. It doesn't even ask me for permission to use my camera on any of the browsers. Anyone know what I have to do?
Reading around I see there's a way to do it with a WebXR plugin.
Idea for Placing a 3D Object on a Real Table
With the Meta Quest 3, I want to place a 3D object on a real table. I'm using Unity and the Meta SDK.
I know Meta provides the MRUK (Mixed Reality Unity Kit), which allows room scanning and semantic labeling of real-world objects, but I'd like to achieve this without relying on that step.
Does anyone have a cool idea on how to do it?
What Iāve built so far lets the user define four points (the tableās vertices) by pinching with their Index and Thumb. My 3D object then spawns at the center of these points.
It works fine, but it's quite difficult to be very precise. Iād love to hear if anyone has ideas for improving this process.
Why do you want to reinvent the wheel, I am wondering?
Because the step of 3D scan, draw a table and select the right semantic label is not friendly for people that never use a headset.
Plus, if someone didn't use the right label the application doesn't work:
I think create an application that depend from another to play isn't a great experience for the users
You should automate that steps, as the sdk will detect and then give the user feedback where they could place stuff automatically from the AR data. The user would only see a status of scanning and placing best case with an indicator raycasting against that found table for example
I can't get rid of the "scan surfaces" and "tap to place object" instructions I get on my phone. Does anyone know where and how?
I don't need those for what I'm trying to do anyway
As you dont know, where they come from, I assume, they are from a sample package or something?
@nimble fossil I believe so yes. But I'm new to Unity so I have no idea where to find it.
New to unity and start off with AR might be a bit too much to start with. If you dont know, how unity works in general and where to look for what, you might wanna do this first !learn
:teacher: Unity Learn ā
Over 750 hours of free live and on-demand learning content for all levels of experience!
How do i instantiate gameobjects randomly on the generated planes.
In other words what i want is once the planes are detected i want to stop any new plane detection and once user clicks on "done" button i want to randomly spawn n objects in the generated planes. Can i please get some suggestions on how do i do it?
Btw I'm just learning AR right now so sorry if it's a dumb question.
Did you read into the documentation on the ARPlane manager and detection. In the end, you should get meshes and then you can do with meshes, whatever you wanna do. Raycast against it, take the vertices as range and spawn within etc etc
Might help you unterstanding and visualizing
Lot of devs in here, just ask, what you gotta ask
Hey was wondering if anyone is familiar with the Meta XR SDK. Im trying to implement grab and toss and while the grab function works using Meta's building block Grabbable, tossing does not work
Hi, I'm a beginner to Unity and was trying out AR Foundation Pathway. When trying to build and run my AR Experience (Simple Box in my room). There is this yellow background (image attached) that comes instead being able to look at my room. My project is set in URP 3D. I'm using an Android device.
This yellow bg is also present when I test the game on my PC
Did you set the camera background to black (no alpha) and enable the URP AR background render feature?
I think you have enable physics on it somewhere
I'm not sure about this, could you tell me from where can I find these settings?
First is on the camera itself. Set clear mode to color instead of skybox and to black
Second is in the URP renderer
It's interesting that we can't simply use androidXR on Android phones
Is it possible to make Quest 2 use my phone camera instead of its own cameras for a AR game?
I need the colors and Quest 2 doesn't have color cameras
Hello Unity Community,
Weāre facing an issue with accurately placing 3D models at ground level. Since scanning the entire environment isnāt feasible due to the large space, weāve tried using predefined elevation data to position models. However, weāre encountering inconsistenciesāsometimes the models are placed slightly above or below ground level.
Has anyone dealt with a similar issue? What would be the best approach to ensure precise placement without real-time scanning? Would raycasting, terrain sampling, or another method be more reliable?
Any insights or recommendations would be greatly appreciated!
Thanks!
@crystal hedge There's no advertisement on the server.
You could raycast against both ground plane and depth, maybe even across multiple frames, to have a more accurate placement maybe
Definitely not
Your question asks exactly for realtime scanning. Depending on the target device, you can enable/disable specific AR scanned objects like planes to get correct values and create a map from those points, so it does manage the world position. But what platform are we talking about? Because, sometimes it couldjust be a limit of your hardware, to not be more precise than it is
I need help setting my scene for the meta quest 3, and the only thing I wanna apply is night vision using post processing
in xr simulation the image recognition is broken , why is this ?
Hi everyone! I need to create two apps with hand tracking, a trivia and a puzzle game. I was planning to use manomotion sdk for it but i have two doubts
1- if i pay manomotion sdk pro on the asset store i can use it all i want?(i will use the games on events in a screen)
2- can i use it for windows games?
- If it doesn't state any special license, yes. You should be able to use it in all projects for any platform
- For technical questions, ask them directly.
Perfect, thank you!
Error Unity DllNotFoundException: Unable to load DLL 'UnityARCore'. Tried the load the following dynamic libraries: Unable to load dynamic library 'UnityARCore' because of 'Failed to open the requested dynamic library (0x06000000) dlerror() = dlopen failed: library "UnityARCore" not found
Error Unity DllNotFoundException: Unable to load DLL 'arcore_unity_api'. Tried the load the following dynamic libraries: Unable to load dynamic library 'arcore_unity_api' because of 'Failed to open the requested dynamic library (0x06000000) dlerror() = dlopen failed: library "arcore_unity_api" not found
Need help on this. I am building an Unity AR project for Android. I am facing this issue. The app open but there is no camera feed from the Mobile. The mobile supports AR. But Idk how to solve this issue.
- How did you enable AR support?
- Does the phone support ARCore and the required features you selected?
- Does installing ARCore services help?
I have sorted the issue. Thanks!
I had issue with the plugin duplicate loading. Cleared it out!
I'd like to do a colocation app but without internet connection with Unity and Meta Quest 3.
Is possible do it in P2P? And I bet I need a workaround for colocation because SSA doesnt work without internet.
Anyone ever try do it?
Why do you think you need internet connection? not sure its jus tbecause they talk about photon in their docs, but you could use any multiplayer package. Or am I missing that information somewhere
SSA does need internet to sync the id, but thatās an example of automatic co location. The manual style would be to put your controller down in the same position/orientation as the other users and sync based on that point, and that would work without internet.
Between you and me the meta spatial anchors arenāt particularly space accurate so manual sync might be better anyways š
Hi everyone noobie student here ! Do you know which product will be good for a simple pokemon go like with unity 6 ? (Map, POI trigger AR screen, scores) I tried mapbox and niantic but they seem not to be up to date 
Unity 6 is only with AR Foundation.
But if you want to make something like Pokemon Go, use Niantic tools as they have the whole POI/Maps/VPS system ready. Just use Unity 2022.3 and upgrade to 6 when they release their update
Unity 6 support will be added in ARDK 3.12
Thank you for the fast answer
ā¤ļø
We don't have camera access in quests. And I don't think cross communication with another camera would be supported (like in ever)
@warm abyss you practically know everything. How?
Are you Unity discord's secret AI?
Its just a texture stream you need to setup, but latency will make the user dizzy as hell š
Or use sth like image markers for initial setup or visual box drawing. Depends on the usecase
4 years of game development bachelor (specializing in programming + XR) and during the last 6 years (4 of those years + 2 years freelancing) I am active on Discord and the forums. And I read most release notes etc for new versions.
For some reason with new information about game dev my brain is a sponge haha
I wanted to add, I definitely don't know everything! I only wanted to share how I got the knowledge I do
Hey may I know Unity Human Body Tracking support ARcore?
There is no native support afaik, so you gotta find another solution
Yep, confirmed with what Dev Dunk says, for ARcore there is no Human Body Tracking built-in
"active on Discord and forums", My man is literally an unpaid support engineer.
As an alternative, you can use any of the OpenCV examples for this though.
They will take work, the most popular one in the store, I had to rewrite it's world space translation. But we are using it on commercial apps with success.
It also does hand tracking, object identification, etc. It's good to know as an alternative.
hi guys! i was wondering if anyone could help me figure out how to keep my tracking a bit more stable? it seems to drift off for a split second every so often š this is my first project so im very new to all of this, im also using vuforia. My image shows up as 5 stars
I didn't know people still used Vuforia. Maybe AR Foundation would work better?
I know the image tracking there at least is very solid.
oh i didnt know it was older.. do you think itll be hard transfering everything over?
Its not a big deal to use the ARFoundation image library asset instead of vuforia. tons of tutorials on unity and samples. But for your drifting, as I assume, your screen is curved. The tracking expects a flat image. So to be 100% sure if your marker is working, you need it as its expected, on a flat service. The slightest bend can affect the tracking already, even printed on paper and its a bit bend. so be sure to maybe laminate it with matte finish or something if you want best performance on tracking no matter the underlying system#
ar foundation is some of the most awful stuff ive ever worked with what is this
It could also be you're balancing on the edge of the tracking distance for images, does this drift occur when you're a little bit closer? I see you're using vuforia but I'd still be curious to know the answer.
Subjective opinion id say. I am having a good time using it š
I'm trying to access the trackedImageprefab which is spawned by artracked image manager to control it via a joystick
I couldn't find much info about it in the forums
can anybody help me with this pls
nvm solved the problem
Dear visionos people, according to this docs, we can have reflective materials with some shader work https://developer.apple.com/documentation/visionos/enabling-video-reflections-in-an-immersive-environment. Are there any plans to support this in the near future, or would that (for now) need to be a "hack" after your material conversion?
Anyone know a quick and dirty way to do that "wall slowly opening up to reveal virtual objects" effect in meta's sdk? I can't find info on how to remove a specific wall from passthrough
Usually using stencil buffers (using render objects in URP)
was hoping there was an easier method built into the SDK for quick prototyping
The Meta Movement SDK is the closest you'll get, but locks you in. It's a complex system, so there aren't a lot of free options
The Pico 4 Ultra has real body tracking with custom trackers
huh? i was talking about opening up a wall, not avatar movement
Did you even look into it? Asking for an "easier" method just makes me assume you did not as you in the end just use renderobjects rendererfeature to overwrite render params to hide stuff behind others for example. Maybe there is also a meta version of the unity builtin example shader XROccluder, which hides objects. Google for unity xr meta portal will get you several tutorials
Hi! I am attempting to implement image tracking on visionOS using ARTrackedImageManager. I have configured multiple images in the XRReferenceImageLibrary, but only the first image in the library is consistently being tracked. If anyone knows how to track multiple images simultaneously on visionOS, I would greatly appreciate your advice.
Yes, I donāt ask questions without checking the docs first, and I am very familiar with them although sometimes we all miss stuff. I said easier because Iāve used renderfeatures and it seems a hackier than expected solution for an effect I see a lot. Trust me, there are much less essential things packaged built into in their SDK
Just to clarify, you mean a detected real wall or a virtual wall (as mask)?
Unity Version
- Unity 6000.0.25f1
Packages: - AR Foundation 6.0.5
- XR Simulation Environments 2.0.1
- XR Plugin Management 4.5.0
Scene Hierarchy - Defer to screenshots
Runtime - Defer to screenshots
Tutorials Used - https://docs.unity3d.com/Packages/com.unity.xr.arfoundation@6.1/manual/project-setup/project-setup.html
Problem
- I have been following the Unity Documentation for setting up a basic AR scene, but when I hit play, I get a yellowscreen. I can control the player perfectly in the scene view and see that all the objects are there, but the game view is just a yellow screen.
- XR Simulation also creates a new
SimulationCamerathat is disabled that I don't really understand the purpose of, since it's disabled explicitly.
Potential Problems
- I'm not super good at URP so it might be borked
Update: I forgot to add the scene to the build š¤¦
can you filter for Camera in your hierarchy? I assume, there is some camera writing on top of all others
It sounds like you might not have set up the render pipeline. Check Graphics in Project Settings and make sure you've set a Scriptable Render Pipeline Settings asset at the top, and also that you have a Render Pipeline Asset set under Quality.
Hey Iām trying to detect objects using unity and yolov8n would really appreciate if someone could have a look at setup as I feel like I may have done something wrong because when I build and test the application to detect objects, my Samsung is sending black frames to the server and not actually frames of whatās being shown and so naturally nothing is being detected
Break it into small chunks. Just have a small texture in the corner of your screen that takes in what you read. That makes debugging a LOT easier already.
Otherwise, we need a LOT more info to help
Thanks Iāll try that I think Iām also gonna try to save the frames before theyāre sent to the backend server to see if the problem actually is due to the camera
Hello. I have another AR related problem and I posted it on the forums if any of you guys have an idea... Basically The ARKit's own Coaching Overlay system gets stuck and it never returns a result when the user moves/rotates the iPad around.
how to fix this it is not showing the background using new unity 6 lts and ar template
Guys Iām actually losing my mind pls someone help me im making an Ar hazard detection system to help reduce collisions involving motorcyclists but the frames being captured to be sent to server which does the hazard detection is just black please help
Do you use URP? If so, does it have an URP renderer active with the AR background?
I personally prefer using samples over full templates, so it might be worth it to try some included samples in a project that is known to work
What did you do to read the images?
Which debugging steps did you take?
Are they in the right format for the server?
I just have a method which saves the frames before theyāre sent to the server. The debugging steps I took was adding permissions adjusting pipelines and loads of other stuff Iām at the point where I ran out of idea so Iām js getting ai to modify my code but even thatās not working. This is leading me to believe itās probably something to do with my unity set up
Save the frame you read to a file and see if that works. To see if it's a Unity issue
Of that's black, you're saving the frame wrong
Only using AI to fix bugs without understanding what's happening isn't smart
Yes but I was so very stressed about this
Do u mean like this
Video doesn't load for me. But in the screenshot I don't see an image. Write it to a png or jpg. Don't immediately send it to the server when you don't know if the original image is good enough.
And paste your frame saving code here
Here is the the code that saves the frames
debug_dir = "debug_images"
os.makedirs(debug_dir, exist_ok=True)
import time
debug_filename = os.path.join(debug_dir, f"received_{int(time.time())}.jpg")
with open(debug_filename, "wb") as f:
f.write(file)
logger.info(f"Saved received image to {debug_filename}")
Here is what frames I get
Iāve also started to get this error in the device logs
Texture must have width greater than 0.
UnityEngine.XR.ARFoundation.AR CameraManager.UpdateD
Unity Exception: Falled to create texture because of invalid parameters.
UnityEngine. Texture2D.Internal_Create (UnityEngine. TextureD mono, System.in132 w, System.In132 h, System.Int32 mipCount,
UnityEngine.Experimental.Rendering.GraphicsFormar formal,
UnityEngine. TextureColorSpace colorSpace.
UnityEngIne. Experimental. Rendering. TextureCreationFlags flags, System.IniPu nativeTex, System. Boolean IgnoreMipmapLimit, System.Suing mipmapLimiGroupName) (ai <00000000000000000000000000000000>:0)
UnityEngine. Texture2D.Creale ExternalTexture (System.in132 width, System.in132 height, UnityEngine. TextureFormal formar, System.Boolean mipChain, System.Boolean finear, System.intPu nativeTex) (ai <00000000000000000000000000000000>:0)
UnityEngine.XR.ARFoundation.Updatable Texture2D..ctor (UnityEngine.XR.ARSubsystems.XRTextureDescriptor descriptor) (at <00000000000000000000000000000000>:0)
UnityEngine.XR.ARFoundation.ARTextureinfo..ctor
(UnityEngine.XR.ARSubsystems.XRTextureDescriptor descriptor) (al
<00000000C00000000000000000000000>:0)
UnityEngine.XR.ARFoundation.NoSwapchainStrategy.ResizeTextureinfos
(Unity. Collections.NativeArray'1 (T| descriptors) (al <00000000000000000000000000000000>:0)
UnityEngine.XR.ARFoundation.NoSwapchainStrategy.UnityEngine.XR.ARFoundation.JS
wapchainStrategy. Try Update TextureinfosForFrame (Unity.Collections.NativeArray'1/ textureDescriptors,
Uniry.XR.CoreUtils.Collections.ReadOnlyLisiSpan*1(UnityEngine.XR.ARFaundation.ART
extureinfo]& textureinfos) (at <00000000000000000000000000000000>:0)
UnityEngine.XR.ARFoundation.ARCameraManager.Update 0 (at <00000000000000000000000000000000>:0)
I didnāt have these errors before but now I get them even when I go back to my old code I hope fixing theses can fix the issue
!code pls, even for logs, it can make it easier to read
š Large Code Blocks
Use links to services like:
https://paste.mod.gg/, https://hastebin.skyra.pw/, https://paste.ofcode.org/, https://paste.myst.rs/, https://scriptbin.xyz/
š Inline Code
Surround code with three backquotes. Not quotation marks.
To format as C#, add cs to the first line:
```cs
// Your code here
```
Add a comment with a line number if there is an error message.
And as the error states, somehow your texture is 0 width, so you do not get a correct image. But without the full code AND highlighting, at least I am not gonna try to understand it š
Oki, so I fixed that, but now, when I build to my android, the camera just displays the skybox. I'm not sure if it's just not using my irl phone camera or what.
I'm going to try to just remove the skybox and see what happens
for context, I've just build this basic scene with literally nothing but what I show in this photo, and all it displays is the skybox:
I added this asset to the Quality settings as well as to Graphics. All I did was drag-drop the scriptableObjects in. Build still just displays skybox
idk if this works like this but heres my code:
i have set the hight and width of the texture
A tool for sharing your source code with the world!
sorry did this work?
can anyone confirm if they can see my code on the website?
I need help on Unity Mars, Iāve one query. Could you please help me to solve
My query is : I want to make the object to disappear when camera stop seeing that image marker. (unity mars)
it works
Ty
I'm trying to make an AR game that can recognize hand gestures like grab and lets you interact with AR objects with your own hands, does anybody know how? Also how do I use mediapipe on unity there are like no up to date tutorials
Hi whatās the difference between screen capture getting black frames and getting grey frames cos I went from black to grey
According to Google itās a step in the right direction
Can anyone confirm this?
You can check unity 6 and the multiplayer examples for VR/AR, those are holding some objects that help you getting started
Not really, sorry. Can you reiterate, what you did and what the current problem is? Still the texture being 0?
Yes it is what I done different was move around things in my unity set up
And also go to scene list and add the one I was working on
Unity version: 6.0000.0.25f
AR Foundation version: 6.0.5
I am having an issue with AR Foundations where, on the first time I hit play in the editor on my pc, my project starts fine, but the second time I run it, I get a slew of errors related to XR Management already having an active loader in the scene, while the XRSubsystems don't have an active loader configuration. I think that means that the active loader instance isn't a proper singleton, or the logic for it's singletonness is broken.
I have seen some solutions that require making your own scripts to de-initialize the current active loader instance before making new ones, but isn't there any better solution to fix this? One that preferably uses the functionality that already is implemented.
You might wanna share your errors?
š¤¦āāļø oof I thought I already did. Give me 15 minutes and I'll be able to slap them on the original post
When you launch it for the first time, what is the error you get at the end, and how do you exit the program when you're done testing?
The errors here make me think about times then I've run into a sudden crash or unexpected stop, which causes memory leaks and the domain to not reload, which can lead to the editor saying that you already have everything set up
so I need to implement a quit button into my game and that'll fix it?
when I launch it the first time, no errors occur, and I just exit by stopping the program
Like, with the in-editor buttons
it's subsequent times that I hit play that it happens
Hm.. did you turn off domain reload at any point in your dev work?
No i haven't. I wasn't even aware that you could do something like that lol
Currently it's set to Reload Domain and Scene
I still can't figure out this bug so I'm just gonna report it to Unity as a bug
Are there any plans to remove that need ot videos being in two places on visionos builds? This is just a showstopper taking the limit of app size into account and also build times.
hey i have some errors that i cant seem to fix can someone please help me
im getting this null reference exception when using arcore extensions, i want to use geospatial api for my project. im using unity 6000.0.31f1, arfoundation 6.0.3 and arcore extensions 1.48.0. i added the arcore extensions game object and added everything in the inspector, but im still getting null reference exception.
i really need help on this and i would be really grateful if i can get it to work
Are you trying to run ARSession in editor? There is no ar session available in editor, unless you emulate it somehow
Yes
Where does your error come from when invetigating the stacktrace?
But when i dont have arextensions gameobject i dont get any nullreference errors
Im sorry i dont exactly know what that means
stacktrace is the message on the bottom of your console when you select a message on the top view of the console. That shows blue highlighted, what script is asking for it. Is there any script from you or is it all inside the library package cache scripts?
I think its all inside the library package scripts, i'll recheck again in a while
You can sent a screenshot of it. But mayb eyou need to instantiate that ar extension gameobject after getting the arsession and not load it, if its in editor. Been a long time I used geospatial anchors
Ok, i'll send a screenshot in like 10mins im otw to my lab, also which version did you use?
So like make it a prefab and instantiate from the script?
its at least 2 years ago, so outdated š
Maybe its the version that im using, should i try downgrading
Oh, and do you remember if you used AROrigin or XR origin
@nimble fossil
its all inside the library package scripts
No its not, GetLocation is your script
I think, there was no XR back then, it was all AR
oh yes sorry my bad
but im using ar earth manager in get location thats why
yeah maybe its that?
You just gotta check, if there is an active arsessionĀ || editormode and do not try to access features you cant access then.
wdym by an active arsession or editormode
your error is crying about an arsession not being there == null. So you gotta check, if there is an active arsession at all. Google for that and learn the docs of ARFoundation to know, how the whole system is setup. Maybe there is also example scripts for the extension package you can look into
hmm okay
i'll do that
@nimble fossil ```using System;
using Google.XR.ARCoreExtensions;
using UnityEngine;
using UnityEngine.XR.ARFoundation;
public class DebugARSession : MonoBehaviour
{
public ARSession aRSession;
public GameObject _ARCoreExtensions;
void Start()
{
Debug.Log(ARSession.state);
ARSession.stateChanged += OnStateChanged;
}
void OnStateChanged(ARSessionStateChangedEventArgs aRSessionStateChangedEventArgs)
{
Debug.Log("AR session state changed to: " + aRSessionStateChangedEventArgs.state);
_ARCoreExtensions.SetActive(aRSessionStateChangedEventArgs.state == ARSessionState.SessionTracking);
}
}
i wrote this script to only activate the arcore extensions after my state is "SessionTracking" but im still getting the same errors
@nimble fossil dude the craziest thing just happened
i built it and ran it on my mobile and everything worked
its showing the lat lon and altitude so accurately
Yes, because your build has an ARsession available
yeah im stupid af
ive been stuck on this for the past 3 days and i havent tried building it to my phone
Now you know, unless you are using something like MARS or similar, you cant test AR specific features easily.
You can also use the latest unity ar foundation remote app, you gotta build yourself, so you can stream your device data to editor: https://makaka.org/unity-tutorials/ar-testing
omg
i didnt know
tytyty
ly
this would be really helpful
i wouldnt need to build to test it every time
@nimble fossil i wasnt using the development build until now, when i did that i got the development console and got the same null reference exception. my coordinates are showing on the screen but whenever i add a text and store it to my firebase my app crashes, do you think its because of those null references
Can you tell me how to fix it
Does anyone have any resources for how to set up room recognition/mapping with the Apple Vision Pro? My lab wants me to create a project where I can place AR objects around a room and use the AVP headset we have, but I just cannot find any resources on how to do it. I know that it uses ARKit and I have that installed already. I have developed with unity and AR with vuforia before, but I know that it's different with ARKit, and especially different with the stuff for the AVP
Or any tips for how to align a digital twin mockup in Unity to the real world when using the AVP
you can just use arfoundation and its plane detection manager or other as you need. Hook into their callback events and do what every you like with the information you are given
I have never worked with callback events so I am not sure what that exactly is. Does using arfoundation and plane detection have a way of just automatically aligning to the room so that when I build and start up the app on the apple vision, it kind of just goes without any calibratino/setting up?
I have read about scene understanding and room reconstruction, as well as world tracking anchors, but have no clue how to link it together to say that if I have a particular AR object in my mockup room in unity, when I build to my headset, it will appear in that exact same spot in the real world because the mockup is perfectly aligned with the real room
you use the arfoundation managers to enable the feature sprovided by arkit on visionOS. But you just gotta read a lot of documentation from unity polyspatial stuff and arfoundation on visionos.
yeah, I have been reading a lot of things, especially from the documentation itself, but haven't found much that is usable so far š
I have developed a small game based on a set of tutorials. The tutorials are using Unity 2022. and v68 of Meta core. Mine is Unity 6 and v74 Meta Core. Part of the development is having a Raygun in the right hand and a flash light on the left.
This was working up until a certain point and I dont know when or what happened but now even though these two items are attached LeftHandAnchor and RightHandAnchor they dont attach to the controllers.
Here is a screen recording of what is happening in game.
https://youtube.com/shorts/A5SjAeRQig8?feature=share
Can anyone suggest why this might be happening
Here is the hierarchy of the project
I am not familar with meta core, but can you show some code, what should work but does not?
Or is it all setup using premade components
the flashlight is supposed to be anchored to the left controller and the gun to the right. Its currnetly appearing on the floor
premade i suppose
maybe try to track the position of those objects, probably they are not used at all by the system anymore?
maybe its also not using the parent but one of the child objects
firstly thanks for even trying to answer. Its very hard to get support from anywhere...all i can think of is that I have added something to another component that is breaking the anchors
and its not a comile error, its in runtime. But because I am testing in my Quest 3 i dont see the output
how can i output a log file?
You can attach to the application logging and just output on a world ui for you: https://docs.unity3d.com/6000.0/Documentation/ScriptReference/Application-logMessageReceived.html
I would try a simple scene. is ther elike a raw prefab from meta sdk, that you can just drag drop and see, if it works? If so, try to add a simple cube or something to the hands. go from there
Would I add it as a component to something
what, the logging? Just make a simple UI with a textfield and set .text on it.
yeah the logging. Can I run a script that takes all the logs and outputs it to a file?
did you check the docs I have sent, there is an example script already there. Just put that anywhere in your scene gameobject, set a public inspectorfield for your textfield, drag it and set .text
okay I get it
hey guys i have built an indoor navigation project using google ar core and ar foundation . I have a simple interface which shows the destinations where a user can go and accordingly , it gives the shortest navigable path . now I want to make it a voice recognition app , where in the user automatically says the destination and it should show that path . Any idea on how to implement this?
I'm having problems with AR Foundation again. I'm trying to get horizontal planes in my game to display. It works perfectly in XR Simulation in-editor, but the planes aren't updating their position relative to the player and aren't in the right place on my Android build. I've set the xr origin height offset to 0 and had it not specify the offset
Setting the xr origin offset to 0 might be an issue but the planes were STILL static before I made that change
uhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh
so, I have 0 clue why the XR Origin is visually moving ONLY after I spawn a ball in
Like I can't even venture a guess. It's the root transform on the object. It, by all accounts, shouldn't be able to move, but SOMETHING that it's parented to(????) is causing it to move
has anyone had issues with the line renderer?
you have visual tansform handle set the "center" not "pivot"
center will make it average the position of all children objects, it's not moving, but it's children are I assume
normally I set this to "pivot" and "local" because it makes everything make much more sense when debugging positions and rotations
Woaaaaaah I figured the ball transform was doing something to it but I didn't know THAT could happen lol
Looking for some help. When I run my XR app in play mode there are no errors. WHen I run it on my Quest 3 it crashes after a second. I have attempted to add a logger. Here is loggin script. But I dont see logs appear anywhere.
Can someone advise the berst approach here
Pastebin.com is the number one paste tool since 2002. Pastebin is a website where you can store text online for a set period of time.
You are trying to create a directory in Application.dataPath, which is read-only, I think that may be causing your crash
@full cradle what should I use instead?
Application.persistentDataPath is what you should use
hey has anyone been able to get meta avatars to properly work in their games? Mine keep getting stuck t-posing at 0,0
hey guys
is there any way to get testing of runtime without having to build to my phone everytime?
Except for this: https://assetstore.unity.com/packages/tools/utilities/ar-foundation-remote-2-0-201106 50 billion dollar solution ? :>
I believe Niantic's ARDK has it, which is built upon ARF
Might be worth a shot. But it's not natively supported. I believe it's on the roadmap, so maybe give your opinion there to speed it up
With in-editor testing being better I find it to be quite nice nowadays. But yeah remote testing should be speed up
But I am also privileged with a fast PC for builds
does anyone know how I can fix this error?
Huh that's weird. What is your project setup, like are you using arf samples or did you just import the interaction samples into your own project?
i fixed it, thanks
Glad to hear you got it taken care of!
Please don't cross post, it's enough to post in this channel. I'm not aware of any tools that can do this apart from OpenCV which lets you do the heavy lifting yourself
@hybrid perch Thanks for the heads-up!
Iāll make sure to keep it within this channel from now on.
Regarding OpenCV ā Iām open to trying it out. Has anyone here successfully used OpenCV (maybe with Unity) to detect or track feet or foot landmarks? Even just a rough direction or resources to get started would really help me out.
Appreciate your time!
I'm currently trying to make an XR experience in Unity. Right now, I have an app that uses Meta Building blocks to scan a scene. How do I set it up in such a way where certain objects (such as a table or bed) are properly identified?
anyone know how i would go about creating an AR soccerball that is kickable with my own feet?
They have a prefab that gives the MRUK room details. Here you can go over all objects in the room and check their tags
Put controllers on your feet or get a body tracking solution
im using image targets to spawn prefabs on top of an image, if I want to place a prefab directly on top of the image and leave it there even if the target image is out of vision how can I do it? Right now whenever the target image is not visible, the prefab slowly drifts away
I've done this before and essentially I just reparented the spawned prefab so it is no longer a child of the image spawn prefab
cool cool, you set the parent to null, right?
the problem im facing is the prefab slowly drifts away when I move/shake the camera
Yeah poor tracking will lead to that
In my testing iOS was significantly more stable than Android
I did implement a feature where if you scan the image again after the prefab has been spawned, it will recenter itself
what I want is for the prefab to jsut remain in the same position, I don't want to continually track the image for the prefab to be there
just spawn the target in position X and just leave it there, no matter where the camera is pointing at
That's what I'm talking about as well. But you can rescan the image to move the object back if it has drifted
Essentially with drifting you're limited by the capablities of ARFoundation
hey does anyone know how to make a collider for passthrough? I'm using building blocks to generate a ovrpassthrough layer, but I can't get a rigidbody. to properly prevent objects from falling through the floor and other objects
hi guys! i'm detecting the horizontal plane for my AR. is there any way i can improve the detection since it somewhat bleeds into vertical walls when detecting?
does any one here have experience working on AR apps on an iphone
I am having trouble getting my simple application to even begin to start plane tracking
it simply starts the app, gets through the splash screen, shows the camera for a couple of seconds, then crashes
with this in the log
which basically tells me nothing
if anyone van help me with this I'd greatly appreciate it
p.s.: these problems don't appear when testing on android, all my testing on android have been essentially flawless
guys can you help me?.. I have set the object to be in front of the camera, but when I try the marker the object moves behind the camera.
does anyone here know tf is going on? im returning to the original MR template but it doesnt work anymore???? i need it because apparently you cant use building blocks to: have a cube with physics and have it use the room scan - also the room scan block sometimes displays sideways room scans so i need this template - pretty sure right channel
ok now im gonna try to do MR Multiplayer (i was using unity 2022)
also - do i need the android build tools for vr
if i have unity 2022 with it?
not even LOGCAT HELPS
its going at 50 messages a second basically
please help nothing is working ai doesnt help either it keeps going in an circle
why does this happen I DIDNT MODIFY ANYTHING
Ai can only help you with a coding problem if you already understand the coding problem, due to how it's going to get small things wrong that add up to nothing working unless you know the catches, we need more specific directions or errors saying what happens when you try to run the app or project.
hi guys i need some help in unity AR mobile app , with the default XR template .. i want to create an app to view my model and rotate and scale basically , drag drop ,tap to place .. which is all there in the template , so i just tried to put my own models into those , but they are bit broken , even after replication all the prefab varients and assigning them the control the , interactions are not working like the default once ...
Hi Guys, Does anyone knew how to make Virtual Try of ring in AR using unity? Please no paid sdks.
Has anyone successfully implemented a WebView in VR using Unity? Iām looking for a solution that allows interaction (clicks, scrolling, typing) within the VR headset. Any plugins or approaches you recommend that work well with VR input?
We use Vuplex for a web based ui in VR and desktop, and it works well for us
Thank you I will try it
Is there any free plugin ?
As in the controls aren't working or what do you mean by interactions?
I can't understand what you mean, can you elaborate or try saying it again?
I think there is some web VR stuff you can make but I don't recall the names at the moment. Web VR might be what you're looking for in search terms though
by that i mean , the drag, pinch sclae, twist rotate
but nvm , i went back to unity 2022 lts
and used lean touch for my project
and its done .. and working
BTW is there a way to fasten the floor detection other than hardware changes maybe.. like sometime it detects in second and sometimes it take 10-20 seconds
on which platform?
Guys need ideas for making a project for patients with dementia vr or ar or mr any ideas
Hi ! I want to create a Virtual Try-On experience for rings using AR in Unity. Basically, the camera should detect the user's hand and place a 3D ring model on their finger in real-time. I'm looking for free tools or SDKs ā no paid ones. Can you guide me on how to do this?
I believe there is a hand tracking sample for unity you can find somewhere that may help you with seeing how detecting hands works. After that it's relatively easy to make rings attach to the positions of the finger joints
I want Android but it does support only iphone for finger tracking or hand tracking @true stag
Oh android hand tracking? I'm not sure then, you may need to find something open source that allows hand tracking if the sample provided does not support Android phones.
There is one named Mediapipe by Google but unable to understand how to use it for my custom 3d objects
is there any decent software/site that can check how good my image target is? rn im using pictarize but it doesnt seem to be the most accurate
I've used the vurofia portal for this before, it's years ago though so I'm not sure what's free
I found it, its still free, thanks
I only know of paid ones, we are using a modified OpenCV for hand and body tracking, as Unity doesn't do much out of the box.
But opencv works great for us once you get used to it.
We did a lot of work in that space in the past. But my advice is to partner with doctors/careproviders and build what they are looking for, for us having a medical panel was super valuable. Building something without them is going to be challenging.
hey guys how do i optimize or fix plane detection spilling onto vertical walls?
i've already set it to horizontal detection mode, but the plane generated is not accurate
Guys suggest some material for ARcore with unreal..... I'm using unity from past year for making ar projects
.... Now want to shift to unreal but haven't found any good materials for learning AR
Maybe ask in the Unreal servers
I'm setting up some XR stuff in unity and keep getting this error, i am using the URP, but i cannot find any mengtion of SSAO anywhere. Anyone know anything about this??
i need help i followed a tutorial how to make an ar project the one where if theres a picture it gets show the 3d object well i followed it and i notice something different first the tutorial (aka https://www.youtube.com/watch?v=RKqnzIG4EsI) first he has a material module on the simulated tracked image while mine doesnt so when i put it into the simulated track image script the picture that i want it didnt visually shown both physically and simulated when i tried to play it so i got confused and someone help me i gotta finish this project for my collage
Using Unity 6 to get started with AR Augmented Reality
šš»Need to hide objects? https://youtu.be/TWY1oepHwis
A complete run through of setting up and AR project in Unity 6 using the Unity Builtin pipeline.
Fixing the yellow screen.
Needed Packages
com.unity.xr.arfoundation 6.0.3
com.unity.xr.arcore 6.0.3
Code can be downloaded f...
can anyone assist me
So I'm currently working on an AR game where i have mainmenu scene and then actual game scene. Game scene has reload and back to main menu buttons, on clicking on any of these buttons missing reference error shows up.
The error says:
MissingReferenceException: The object of type 'Camera' has been destroyed but you are still teying to access it.
Can anyone please help me to do it correctly? This is my first AR project so I'm not aware of any clean way to do it.
I need help in to develop Snapchat like app need 1-2 filters with synced facial expressions for 3D model
I'll pay for it
Does anyone know if there's a way to tell/force Vuforia to use the iPhone Pro normal or super-wide camera?
Because by default it uses the wide camera, which can't focus on close objects, making the whole thing useless for small markers.
This is however an android tool, so may not be a "drop in and see the quality" score tool, however I believe AR Foundation does also check the score of images when you make a build with an image library, so you may be able to find some use in that if possible?
unity ofc
arcore, arkit, etc, is different platforms within ARFoundation in unity
Anyone know with VisionOS and Polyspatial, if there is a way to make it impossible to close a bounded volume? Maybe the correct phrasing is 'backgrounded'
Hello! If I have a 2d android game in Unity is there a possibility to launch it on Oculus 3 with AR on something like Virtual Screen? By virtual screen I mean something like this: https://www.youtube.com/watch?v=i_YegFqKSKc
If yes, what kind of tools I can use to do it? I read about some basic stuff related to launching Oculus and setting up AR, but didn't find anything about something I need.
I am just demoing the quality of pass-through as well as desktop usability of workrooms. I was playing around pretty much exclusively with hand tracking and was blown away.
Workrooms: https://www.meta.com/en-gb/experiences/2514011888645651
If you like my content, feel free to like and subscribe!
If this convinces you to get a Quest 3, make su...
last i checked, users can install Android games and load them from unknown sources
Thatās good, but the idea is to āmakeā oculus ar game that actually just an android game.
Something like make a scene in Unity ready for VR/AR and launch prepared android game somewhere in this scene. Something it will be kind of oculus ar gameš¤£
Hello, anyone know how to boost GPU Level to level 4 with passtrough enabled using Meta Quest 3 and Unity?
I know there are some apps that can do it despite the documentation tell different
any good tutorial that use unity + vuforia?
I am making a mixed reality chess game using the Meta SDK and I want to use my hands to grab the pawns. For some reason, it seem that it's only grabbable from a very specific part of the pawn and when grabbed, the pawn seems to anchor/move towards the "middle" of my hand (closer to my wrist) and doesnt stay close to my fingers with which I grabbed it. Does anyone have a clue about what's wrong?
can anyone help me with this? I opened a new mobile ar project, and im trying to build and run the sample given by unity to my phone but i keep getting these errors... i have tried reinstalling the unity editor (unity 6) but it did not fix anything (my phone has usb debugging mode on) im not sure where to go from here. im able to click build and transfer the apk file to my phone tho
Hello I am hoping someone can point me to a tutorial or something regarding a specific project Iām working on. I have already done some work building structures in blender and then importing them to work with VR in Unity. Now Iād like to build a structure in order to have it render in a large flat area utilizing the real life ground and pass through. I figured AR is the way to go but Iām struggling. Eventually Iād like to have NPCs walking around too but on the real life ground. Anyone got a good tutorial I could check out to get this started?
Iām using a meta quest 3, modeling in blender, and running everything through Unity
What is the best approach for sending haptics to a controller? most of what i find is related to depricated components
has anyone here used Manomotion hand tracking sdk? It's not free anymore :<
Set it up according to Meta's documentation for passthrough OR according to Unity XR Interaction Toolkit with AR Foundation
Then enable plane detection or check y = 0, as the quest should have the physical ground at 0
Then place your stuff there
Or the underlying code, not sure what exactly you need
Share the error
I found a good way of doing it š
hye , anyone specialized in using the spatial anchor? Just want to confirm . Is it possible to create an app where it can save up to 4 room with its own spatial anchor. Im using meta quest 3.
Does anyone have any info/resources on how to have persistent world anchors using Unity (no xcode) for developing for the apple vision pro. I do have polyspatial, I just can't find anything about persistent world anchors specifically for unity for visionOS.
Hi, is there anyone I can talk to about the Unity Pro subscription requirement for developing on Apple Vision Pro? (posting in both AR and VR channels).
Just ask your questions here or on the forums
More of an issue than a question. I'm not happy with the $200 a month (per seat) cost of a Unity Pro subscription to develop on the Apple Vision Pro and think it should be based on revenue, as is every other platform, especially considering that VisionOS has a very narrow (and not growing) market share. If others are interested, I'd be happy to volunteer my time and diplomacy assertiveness to escalate this issue internally within Unity. If there is any interest from others, I'd like to know and will start a petition and additional outreach.
Best to post on the forums for stuff like this, but it will be hard to get through to Unity
Still looking for info regarding persistent anchors for visionOS while working with Unity. I know ar foundation/kit has them, but it wasn't available for visionOS for awhile. Thought I read that it is now available but I can't really find much info
Let me know if this doesn't help!
Thanks for sending this. Unfortunately, ut doesn't seem to be compatible with VisionOS yet :/
I found this on the unity discussions, they are saying to pull in the samples for vision OS, but this mentions something called "world anchor", perhaps it's not the correct thing you're looking for? https://discussions.unity.com/t/how-to-implement-persistent-anchors-in-visionpro/1506409/16 Let me know
Yeah, unfortunately I have seen this post already and it's close but not quite to what we need. World anchors are a part of it, but they need to persist beyond application restarts and such, which according to the Unity documentation, it is just not supported yet for visionOS. We are looking into some alternatives, but thank you!
I'd also be cautious. If what you are building isn't purely a game and for say a school, then you also can't use a pro license and need to pay 3x as much for an Industry license.
Thanks. Our apps are definitely not industry related. Both are in the app store now and Unity has seen them.
No worries, as long as they are games you are fine. If they are non-games, your revenue is calculated based on your biggest customer or for whom you built it for.
If it's a violation, you likely won't get locked out of your accounts and forced to upgrade until a sales rep comes across it.
(They are primarily paid on commission, so they have a strong incentive to force payment)
FWIW: I haven't had a good experience with several of the sales reps I've interacted with. My most recent experience was so bad, I asked for a replacement, which has still not happened and it's been about two weeks.
Yeah afaik the VP of sales hasn't changed and was one of the champions of everything horrible with Unity licensing.
All I have to do is wait, tho. Every time I talk to a sales rep, it's someone different, so there's obviously a lot of turnover in that department. The last guy told me he normally worked with enterprise level clients and couldn't answer a single question.
makes sense.
Yeah they mmm were poaching them from autodesk for quite a while, with a high promise on commission. So you get what you get with that.
hi guys, does anyone has experience using XR simulator in Unity? im new to all of this and i'm having some trouble with a project that uses the XR Interaction Toolkit . Could really use some help, Thanks in advance!
I'm trying to have a wall/floor mesh i create to be on my walls/floors, but i also want my meshing geometry that isn't near the walls/floors to occlude the walls/floors so people don't walk into chairs or whatever. What material shenanigans should I do to mask the ar occlusion material out when it's near the walls/floors?
Hey guys, i have a slight problem with the MRUK. Im trying to scan everytime i enter the game im programming using OVRScene.RequestSpaceSetup(); (using Meta Quest 3). However i dont understand how i can save or import the scan as a prefab so my MRUK Script can load it. Is there an wasier way maybe? And if not, how can i always scan the room and afterwards load the room?
Hey, I am somewhat new to Unity and currently following the mobile ar pathway. I had to scale down the beachside town quite far and now I'm at the spot where the tutorial wants me to add a world space canvas. However if I scale it down to the same size as the model itself all of the fonts start just becoming triangles when viewed from a close distance. I realize it is probably to the extremely small scale but I don't know any other way to get it to a size approriate to the 3D asset that it is supposed to be rendered above.
Does anyone know how to get a full xr origin AR prefab that has controllers, similar to how the xr interaction toolkit samples have a fully built xrorigin vr rig? The AR samples from the toolkit has an AR rig, but it does not have controllers
I believe Unity has some MR samples on github. Maybe check those out
Instead of scaling it down so much, keep the scale between 0.1 and 2 and just reduce the size. (width and height)
What part exactly do you get stuck? Only the writing to disk part?
They do, but i feel like I should have to clone a whole github repo just to get the prefab, and I can't just export the prefab out because there are a bunch of different scripts that can't be cleanly exported with it š¤
Poke at the code and learn then I guess
New here how to remove this error
No active UnityEngine.XR.ARSubsystems.XRFaceSubsystem is available. This feature is either not supported on the current platform, or you may need to enable a provider in <b>Project Settings</b> > <b>XR Plug-in Management</b>.
UnityEngine.XR.ARFoundation.ARTrackableManager`5<UnityEngine.XR.ARSubsystems.XRFaceSubsystem, UnityEngine.XR.ARSubsystems.XRFaceSubsystemDescriptor, UnityEngine.XR.ARSubsystems.XRFaceSubsystem/Provider, UnityEngine.XR.ARSubsystems.XRFace, UnityEngine.XR.ARFoundation.ARFace>:OnEnable () (at ./Library/PackageCache/com.unity.xr.arfoundation@ddd5a77596e0/Runtime/ARFoundation/ARTrackableManager.cs:98)
What component or object that can work like AR placement Interactable in XR Intreactable tool kit 3
Raycasts?
I'm talking about api I did with raycast but thanks.
Im having trouble with an imported .fbx no appearing in the render. Id like a structure i made in blender to appear in the landscape when using a quest 3 in an open area. I have imported it as an fbx put it into the scene, made sure it was the proper scale, and textured it. It shows up in the project view but when i hit "build and run" it doe snto show up in the headset. I am running unity version 2022.3.38f1. Any suggestions?
so apparently it was the "scene to build" setting. I dont understand why the ONLY SCENE IN MY HIERARCHY wouldnt be the one to build. Thats such a dumb oversite
I want to ask some qustion about AR in unity so if someone here who want to help me im glad to know that!
Can we talk in dm?
no
just ask in chat, people dont do DM help here. if your issue requires that much space in chat then make it a thread
Hey guys!
I'm trying to make a project kind of like the one in this video. From what I can tell, it looks like it uses computer vision with AR in Unity ā is that right? If it is, can someone explain how AR is linked to computer vision?
Harness the potential of Augmented Reality for rapid and lasting value, AR use cases for service and support are easy to deploy and scale hence you can provide increased value to your customers and Service Engineers.
Hey anyone who worked on AR world tracking using unity and AR foundation?
Does anyone know a way to identify a room using MRUK and Quest 3? Like I want a specific version of an object to spawn in a specific room between the ones I have scanned.
I've tried pooling through MRUK.Instance.Rooms, but it doesn't always return the rooms in the same exact order.
Anyone using Photon Fusion with VisionOS on Unity 6? I'm getting a nanosockets error. Looks like they built the library for the simulator, but not for the device.
Yes, I'm sure many many others have as well.
a workaround for this: make markings on a wall inside quest 3's room scan configuration and mark them as wall art (or windows) and then make unity count how many anchors of that type a room has, and identify the rooms that way (like setting room #1 to have 2 wall art anchors, room #2 to have 3, etc.)
would be easier if meta just let us access the name we gave to the room inside the configuration, but alas
I'm looking for some outdoor cameras that work well with unity for human detection (or tbh could just be thermal blob detection) to be used in lowlight situation with unity.
does anyone have any suggestions?
range around 30 M if possible, will probably use 2 in order to get some stereo vision if required to do depth estimation. but also, if its thermal, i might be able to do markers in the space to assist with depth detection to
Hey guys, does any one had some experience with color mapping where materials behave like transparent even though i set them as opaque. The project is using AIRAR Color Mapping with Vuforia.
Is not even flipped normals, I've tried to flip them, set render faces to both. One good case was setting different materials in which the sorting priority was higher for the toppings and cheese, but in cases like the lady and painting frame, this strategy has no use as from the side view we can see the painting background as it was failing depth, but i've enabled depth write in another material and same results
Was thinking, maybe it's some problems with the plugin library, so i'm creating a branch to see with ARFoundation, but idk, seems like something not related. I have a small understanding on shading, but not enough to understand what is happening in this case.
Can you show your shader for the pyramids for example?
this is the shader from the box behind the object
this is the setting of elements from scene: camera, object and box
about the shader used in the object, i've tried many things, opaque/transparent, renderface(cull) of all types, even change sorting priority.
If I want to create a mobile app that uses 2d ui for navigation but also utilises ar in certain features, how would I go about setting this up
Which can of setting?
Project Settings?
ARFoundation?
hey guys, I have published an AR app on Google Play that aims to teach computer networks. I did it as a teacher to help my students better understand the basics and visualize it better. I have achived good results on them, but they havent given me much feedback to where things can improve and how I can take it up a notch. Is this a good server to post my app and maybe for someone to try it out?
Wdym?
Assuming it's not a direct download (since we can't verify stuff like that), a Google Play app should be fine if you tell people where to find it.
How do i create just a simple tester scene where i map an image above a users face. Trying to do it from my PC and I just get a black screen
Whatās the status of PolySpatial? Is it still actively developed and supported?
@hybrid perch any help? š
Hi, I donāt understand the effect you want to accomplish with the shader. Do you have any reference or example so I can see if I can help you?
in this image you can se that the painting frame seems transparent, as the internal part of the model is displayed over frame for example.
Did you try to change to Render Both Side?
yes, it becomes uglier as if both faces we're fighting for priority
i'll try to send an example
I can't really begin to debug custom shaders. I'd have a try with the standard shader
I'm currently finishing a build for the app stores that will come with this error, here is night time and really hard to make a test for renderface both. but tomorrow morning i'll mention back here š Thanks so far for the help š¤
there you go. the first image is with only front face, the second one is both faces, and the third is how the default model is in blender (notice that there are no normal miscalculations), forth is in the unity import preview
Great! So if anybody interested to try my app out that teaches computer networks I would much appreciate any feedback āŗ
https://play.google.com/store/apps/details?id=com.CriticalGlitch.ARnet&hl=en
Does it run on Hololens 2?
Made it purely for Android phones. It would need to be build for windows to support hololens - but no plans to do that for now.
Hello everyone,
Iām new to Unity and currently working on developing XR games for the Meta Quest 3 as part of my PhD project, focused on older adults. These games need to be mixed reality experiences, where users remain aware of and can still see their surroundings.
Iād really appreciate your opinions on which Unity version I should use: Unity 2022 or Unity 6. What are the main pros and cons of each in the context of XR development for Quest 3?
Thank you in advance for your insights!
Unity 6 has all the latest features, so use that. 2022 is EoL and doesn't get any more bugfixes as well
ok, will do, thank you. My hesitation with unity 6 was due to the fact that I've heard it still very unstable.
That was a long time ago. 2022 has the same reputation when it came out. 6 has been out for a while and got a ton of bug fixes, so I personally recommend using it
6.0 that is, I wouldn't use 6.2 atm
ok thank you so much!
I recently participated in an XR hackathon, which was my first real experience exploring the XR side of Unity (though Iāve been developing games for about 2 years). I really enjoyed building for VR, and it got me thinking about shifting my focus more toward XR development.
For now, Iām particularly interested in creating AR applications to support learning in secondary schools and under-equipped universities here in West Africa, since many institutions donāt have access to proper learning equipment.
Does it make sense to switch my focus to XR development, starting with AR? And looking ahead, what does the future of AR look like ā will it be sustainable as a career path?
For the future, that's hard to say with certainty more than most companies would say, but as for advice I'd say to look up what people have done with augmented reality things to perhaps give you inspiration. I've seen someone use a quest headset to help draw out an exact copy of an image on a wall-sized mural scale, which I thought was great.
I think a valid concern with regards to the project you're describing is whether the hardware is capable of running AR. A lot of older or cheaper Android models do not have support for Unitys AR Foundation. So if you're considering going in this direction with such a goal in mind, it might be worth doing some research into the potential for implementation.
Something else you could consider doing is get familiar with the Sketchfab application. They have a lot of models made for learning, like anatomy, natural sciences and more. And they have an AR display function. So it could be a way to trial run this tech as a learning tool
Ohh yeah thatās true, Iāll start with that
Thanks
Hi everyone 
I donāt have much experience with augmented reality ā I mainly work in VR, and only briefly touched on XR. I wanted to ask: is it possible in Unity to create an app for Meta Quest headsets that can recognize real-world objects and visualize them for the user? I hope someone has any experience with this problem.
I know developers donāt have access to raw camera data on Quest, so Iām wondering if something like this could be done on the shader layer or in another way. My idea is to create my own project where I implement custom machine learning for object recognition. I know the XR SDK provides some functionality in this area, but Iād like to go deeper into AR/XR on my own, and I think this project would be very interesting.
Hello,
For Meta Quest 3 and 3S, you can access the image through WebCamTexture.
So yes, itās possible to use the image to feed your ML model and track objects.
uuu thank you.
Just look for Passthrough Camera API
You donāt get much for free in this realm, but check out OpenCV and how it can be used in Unity
Hi everyone, is there an an easy guide on how to upgrade a project vuforia project unity 5 to any running project for unity 6?
Tbh this depends too much on the project. Afaik the way is to upgrade the project, update vuforia, check the upgrade guide in the docs, and fix issues bit by bit
can we track arbitrary objects in meta xr? ex: a car
No, you'd have to use something like OpenCV
very cool that we now have camera stream access
You have to use your own model. With access to the camera, it would be possible to integrate it using Unity's Inference Engine and a pre-trained ONNX model such as Yolo:
https://docs.unity3d.com/Packages/com.unity.ai.inference@2.3/manual/index.html
Hello,
I have never used Unity before but I am doing an AR App for a book, that's supposed to track images and play videos on top. With a few tutorials I was able to use AR Foundation to track the images and have it play the videos. Though I was wondering how I can add a stop/play button and a volume slider to the prefab, so that it also pops up when the image is tracked. I tried with tutorials but I've been getting nowhere xd What I was trying to do was adding a button component to the quad where I am playing the video, is that what I'm supposed to do?
Thx in advance
You're going to need to look up some more unity tutorials, this will have to do with unity world-space UI and canvases, then making sure they are clickable and making sure that click will execute code that you wrote to do whatever you like.
Hello guys I have a question can we use vuforia and webXR extension for image tracking and host on any platform ??
What exactly do you mean? Host what, or build for any platform? You will find more info about that on the vuforia page. But be prepared, that you have to check on licenses and that you are tied to this specific package then.
anyone here work AR with computer vision ?
has anyone worked with image and object tracking in the newer versions of unity? I just bought a new mac book for faster development but I can't build my ar app to my phone. I keep getting an error related to actool. It works on my old mac book probably because I havent updated my OS
I recently purchased a Quest 3S and Unity does not see it. It is in developer mode, I am using a USB 3 cable (passes the cable test) and the Meta Quest Link sees the unit. I'm at at total loss here. I have uninstalled all Unity versions and the Unity hub and re-installed, same with Meta Quest Link. The strange part is that Unity will see my old Oculus Quest 2 fine. I hope I'm just missing something simple. Any help is greatly appreciated.
Hi there, has anyone been using Maxst for their android AR projects? Google play has a new requirement with the 16 KB Page size. I'm using the latest version of maxst for multiple (on production, and live) projects and Google play is saying Maxst does not meet that requirement. I have tried contacting Maxst and posted in their forums but no response yet for a month now. Curious if anyone has encountered it and have had any luck solving it or maybe had a response from Maxst.
Is image tracking possible in web through unity ?
hey can someone help me with a problem I have with Unity and Vuforia with area targets ? Im using Unity 6000.0.57f1, I followed every step of videos about area targets, but it seems that when I open the app on my phone it doesnt work, first it didnt open my camera, then I erased vulkan from Graphics API settings on the project settings, then it finally opened the camera but when im at the palce of test it doesnt recongnized nothing from the scanned area target, what am i doing wrong ?
Im using an Honor 200 Pro for the test and i used an Iphone 14Pro Max i borrowed from a colleague
When refreshing or plugging it in, do you give dev access in the headset?
If you can get the webcam, of course!
Just not using AR Foundation. WebAR isn't supported yet
Thank you today I have look a plugin or package known as needle do you have any idea about it ??
OpenCV is the more complete CV package afaik
De Panther on git has the WebXR exporter. No idea if that has AR support tho
You are right de panther has only one problem they dont support image tracking if I am not wrong.
Hello,
If you want to keep using Unity, you can try Needle Engine, which supports Image Tracking for WebXR. If it doesnāt meet your expectations, I would recommend using Niantic Studio.
Why didnāt anyone reply to me?
Hi,
What do you need?
I have a friend who is working on a computer vision code in Python. My question is: how can I connect this code to Unity so that I can build an AR project?
The idea is that when I open the camera, I face a problem, and I want AR to help me solve it.
First, I want to know how to connect Python code with Unity, and then I want to know how to actually start working on the project after connecting them.
Hi people. I have percilour problem and I am on my edge with this bs.
So, I have AR application (I wasn't the that originally built it, but now I am its maintainer).
And until couple weeks ago it worked totally fine.
The only thing I changed in that time is that I increased Max API level to 15.
Now, when it scans target image, it doesn't show info about it.
BUT ONLY on Android build.
In Unity Editor or iPhone it works totally fine, which makes zero fkin sense.
I am not getting any specific errors or something, at least not in Unity console.
Although, in adb logcat this error occurs few times? Could be this related?
Aslo, I was bit lying, there are some errors in console after all.
I will try needle engine
anyone??
You're describing an incredibly complex use case, well beyond normal questions in this channel. What's your background? There might be better options than Unity if you're gonna run a custom solution anyway
Agreed with Ole. What platform are you targeting in the first place? A bit more info about your infrastructure/setup might be useful too. Connect python code with Unity can be anything from calling API calls from unity to integrating different frameworks into one solution for ios or android and what not.
I donāt know how to use APIs, and hereās a video that explains the idea more clearly. I want to do something similar to that. https://www.youtube.com/watch?v=fxNhU-gLh-k
Harness the potential of Augmented Reality for rapid and lasting value, AR use cases for service and support are easy to deploy and scale hence you can provide increased value to your customers and Service Engineers.
When you do not know how to use APIs at all, you might dig into a project thats way above your experience... There are a lot of things going on in this video, just from image recognition. I guess this is not done with simple object or image tracking but actual AI camera feed detection.
I know this isnāt a simple project ā thatās exactly why Iām asking if anyone here can help me understand the steps to build something like this. Also, could you tell me how to use an API?
You can google, to actually know what API is and then check, if this is the correct thing for you to work with.
This channel is for asking questions about AR functionality in Unity. There is nobody here who would be up for guiding you through the A-Z development of a really complex application, which honestly sounds way overscoped
Hi,
Iām trying to implement a custom behavior for an XR Grab Interactable:
The object should always follow its parentās transform, even while grabbed.
The user should only be able to manipulate the local X position relative to the parent.
Rotation and other axes should remain locked to the parent.
I attempted a custom script inheriting from XRBaseGrabTransformer like this:
using UnityEngine;
using UnityEngine.XR.Interaction.Toolkit;
using UnityEngine.XR.Interaction.Toolkit.Transformers;
public class ParentRelativeXConstraints : XRBaseGrabTransformer
{
private Vector3 localStartPos;
private Transform parent;
public override void OnGrab(XRGrabInteractable grabInteractable)
{
parent = grabInteractable.transform.parent;
localStartPos = grabInteractable.transform.localPosition;
}
public override void Process(XRGrabInteractable grabInteractable, XRInteractionUpdateOrder.UpdatePhase updatePhase, ref Pose targetPose, ref Vector3 localScale)
{
if (parent == null) return;
Vector3 worldTarget = grabInteractable.interactorsSelecting[0].GetAttachTransform(grabInteractable).position;
Vector3 localTarget = parent.InverseTransformPoint(worldTarget);
localTarget = new Vector3(localTarget.x, localStartPos.y, localStartPos.z);
grabInteractable.transform.position = parent.TransformPoint(localTarget);
grabInteractable.transform.rotation = parent.rotation;
}
}
However, this didnāt work as expected: when I grab the object and move the parent, the objectās transform still changes after releasing it. Iām unsure which method of XRBaseGrabTransformer I should use or if thereās a better approach to achieve this behavior.
Would appreciate guidance on whether itās possible to implement this fully with a custom transformer, or if thereās another recommended way to achieve parent-relative grabbing along a single axis.
Thanks!
I'm still confused exactly what you're trying to do, but the following components helped with similar issues in the past. Although admittedly I forget the details. the sdk's grabbable stuff doesn't really play nice with customization from my experience you gotta use a bunch of their poorly documented provided classes and hope it works
https://developers.meta.com/horizon/documentation/unity/unity-isdk-movement-providers/
https://developers.meta.com/horizon/documentation/unity/unity-isdk-grabbable#grabfreetransformer
Iām using openxr libraries not meta
Ah my bad, the verbiage is similar
is there a way i can get the old vuforia 8.1.11 version package?
when i open my project which was working before but now giving these errors
Hey everyone, Iām facing an issue with Unity AR builds on Android tablets.
When I launch the app, the splash screen flashes for a millisecond and then the screen goes black.
If I close Unity (without killing the app) and reopen it, the app works fine and everything runs smoothly.
Iāve tested with an empty Unity project and it works normally, so I suspect itās a configuration issue in my current project.
Has anyone faced something similar or knows which settings could be causing this?
Update Unity, update your ARCore services on the device, update AR Foundation and ARCore if possible
Once all that is done and it still breaks, check for errors in logcat
Weāve confirmed that both AR Foundation and ARCore are up to date, and the device itself is also updated.
Kindly check the video above for more details.
As for the logcat, we are seeing the following error:
"Skipped rendering frame because GfxDevice is in invalid state (device lost)"
Additionally, the splash screen logo is not the Unity logo (even though we havenāt added or customized any logo).
- Also update Unity to the latest bugfixes version
- Are you using OpenGLES3? If not, use that
Thanks, Dylan!
Weāre already using OpenGLES3 only (Vulkan removed).
Unity version is 2022.3.62f1 (LTS).
The problem only happens on Samsung tablets (phones on Android and iPhones are fine). On the tablet the system splash (app icon) shows, then the screen goes black the Unity splash never renders.
From logcat we see:
Unable to open libpenguin.so: dlopen failed: library "libpenguin.so" not found
occasionally: Skipped rendering frame because GfxDevice is in invalid state (device lost)
An empty Unity project with the same settings works on the same device.
Same settings? So using AR as well?
If so, maybe copy over all project files to the original project?
Yes, we already tried copying the entire project files into a new Unity project, but unfortunately the build still faces the exact same issue on the tablet.
Then it's somewhere in the project and it's debugging time.
I guess add features one by one until the issue occurs