Creating a 3dmap file
Hi
How dose one create a 3dmap file to use object tracking on your sdk.
I would like to create a program which will regognise an amazon alexa dot and augment some 3d modals in AR.
Do you provide a conversion program similar to the one you provide for 2d images which takes 3d modals and converts them into 3dmap files?
Any help would be appriciated.
Kind Regards.
Hello
Have you tried the Maxst AR SDK Demo app? With the Demo app, you can test all of the features provided by the Maxst AR SDK. And it can be implemented in the same form using SDK.
In particular, we encourage you to check out the Demo app first so you can try out the 3D training you want.
Here are links to your app.
https://developer.maxst.com/MD/download/demo
Thank you.
- MAXST support team
If I could steal this thread....
I'm hoping you have a solution to create a 3dmap file from an existing 3d file such as an fbx / obj file.
I've used your app (and other sdks versions of them) and they're less than ideal for accuracy / ease of use.
Some AR SDKs can read a file at runtime to match... others have desktop apps that let you import models and textures to generate the files needed.
If no desktop app exists, are there any plans to create one?
Thanks
Hi,
We currently do not provide the ability to recognize / track in the SDK using the 3D Model file (fbx or obj). Currently, we are focusing on SLAM technology and will research / develop cad or fbx to recognize / track the next step.
Best regards,
Sam
SDK Team, MAXST
Well that's disappointing...
The accuracy and capture is proving to be far too problematic to be useful for my needs. Many attempts fail entirely to capture any meaningful or stable data.
Is there a way to get documentation for the 3dmap file format / data structure and requirements, so that I could build my own 3dmap file?
Thanks
Hi,
This documentation describes you how to make a map more precisely while learning an environment.
https://developer.maxst.com/MD/doc/g/slamguide
Sara
SDK Team, MAXST
@david.sliviak, I have the same problem. I have the 3d-model of the environment that I want to be detected and am looking for someways to adjust the feature points captured by the SLAM method. Have you found any solution for it?
Sorry,
I haven't devoted anymore time to this problem since my work has shifted to other projects.
I'm sure at some point I'll get back into the need for this, but time will tell.