Since Magic changed their app certificate signing method, I cant use older OS than 0.98 to develop.
I bought your OpenCVfor Unity in the asset store but I cant run it because your OpenCV lib is not for Lumin 0.98
Dear ExonSoft, a lot of Magic Leap devices are being sold for cheap. Your software would be a good base for a lot of projects,.

First of all, thanks for the great project. I have tried the face detection scene with my webcam successfully and it beautifully works in real time. However, when I tried deploying the project to Magic Leap, the only thing I see is a blank screen. No menu, no nothing. I only need real-time face detection, so that is the only scene I added to the "Scenes in Build".

My LuminOS version is 0.98.10, I am using Unity 2019.2.0f1 and the OpenCV for Unity version is 2.3.8. I checked other issues like "MagicLeapArUcoExample is just a blank screen" but I don't see a solution there. One additional parameter I am questioning is the Minimum API Level which is 4 right now. Is there something I need to do to run a specific scene on Magic Leap other than following the ReadMe file? I would really appreciate it if you could help me. I really need to get this working. Thanks.

I want to record the scene while running the app, however, when I include the commented out parts of MagicLeapFaceDetectionExample, the app crashes after I press the bumper. Is there a way to record a video of what's going on just like Magic Leap's VideoCaptureExample does?

Secondly, as it can be seen in the picture you posted before, there is a zoomed in version of the real world behind the app UI. Is there a way to run face detection without the UI or to get rid of the zoomed in view?

Hello,

I followed the tutorial as it is, but when I deployed it on my Magic Leap device, it just showed a diagonal red line appearing on the screen, and the image was also not being captured after clicking the bumper many times.
Can someone please help me with this?
Thanks!

For the model and config parameters in the MagicLeapDnnExample.cs file, the required file type for the model is specified for Torch (On Line 25: "Path to a binary file of model contains trained weights. It could be a file with extensions .caffemodel (Caffe), .pb (TensorFlow), .t7 or .net (Torch), .weights (Darknet).")

However, it is NOT specified for the config parameter for Torch. (On Line 28: "Path to a text file of model contains network configuration. It could be a file with extensions .prototxt (Caffe), .pbtxt (TensorFlow), .cfg (Darknet).")

What is the required file type for Torch for the config parameter?

Hi, I have successfully deployed the Yolo Detection in Magic Leap, would you please tell me if that is possible to use MobileNet to do the same thing? Thank you.

I followed all the instructions and got MagicLeapImageCaptureExample to work correctly. However, I am having issues with MagicLeapDnnExample. All I see is a blank screen when I run the example, except the white box, which disappears and leaves me with nothing but black screen.

Hi,
I'm trying to run this on my magic Leap. So far I tried the Aruco Tracker example, but only got a blank screen.
Now I tried the Image capture example, but am only getting a white image and the following Error in the console:

EntryPointNotFoundException: Unable to find an entry point names 'core_Mat_n_1Mat_III' in
'opencvforunity'

Any advice on how to solve this or what I'm doing wrong?

Thanks!

The Image Capture examples works find, but in the Aruco Tracker Example I just get a blank screen.
This might be a similar issue as #6

Looking at the Debug Console is raises issues with the Canvas. That in VR mode the Canvas might not be visible. My attempts at changing the Canvas settings to fix it failed so far though

"Enclose the code related to the WebCamTexture class with #if UNITY_EDITOR and #endif." - could you be more specific which scripts should be removed and which parts of scripts should be enclosed?

Currently have errors when attempting to build, tried removing all examples & scripts with "WebCamTexture" but it created even more errors. Thanks for your help in advance.

I was looking for this example for SDK 0.24.1 but seems to be removed in latest release

thank you

As titled, the face detection example pauses after 2 minutes. I am able to resume the capture by pressing the bumber but needing to do so after every 2 minutes is annoying. I thought of the possibility of a memory leak and tried disposing variables like "faces" or "grayMat", however doing so just causes a crash. What might be the reason behind these pauses? If it's a memory leak, what should I do to solve the problem? Thanks.

Hi,
I have successfully tried the real-time object detection example. Now, I wish to make my own object detection model so that Magic Leap One only detects those certain objects. How can I go about doing that?

Hi guys,

I can able to detect the object but it is not showing the name of the object can anyone help me to resolve this issue. I am having no error in my unity and I have used OpenCV 2.3.5.

I have attached the output of my project file in the below link:
https://drive.google.com/file/d/18PkQOTslh-LoSXa2esnQqe1RviK_Rqs-/view?usp=sharing

Hi, I have a conflict when building in Unity 2019.1.6f.1. On their website ML say:

The custom manifest replaces the Privilege checkboxes found in Player Settings - Publishing Settings in MLTP versions of the editor.

Could you update the example so it requests the CameraCapture permission via custom manifest as described on ML Creator's portal? I currently get build error saying that the package

already has a USES declaration! You may not have your own "USES=" statement in the .package file.

Thanks!