Hello,

Unfortunately, there is no way for me to make a determination by reviewing these objects. The state of the feature is that you need to work with the technology to determine if it meets your use cases.

That being said, here are some general observations/considerations:

1. The aforementioned linked document states, "Geometric complexity is key to distinguish an object from other shapes in the environment." Basically, what this is saying is that the objects should have enough unique geometry to be differentiated from others. Looking at the DVI connector, I can see the potential for false-positive detections due to the slight variations of the male/female connectors. The only way to really know is to use the technology and test it out.
2. Mobile camera systems are challenged in macro scenarios where the camera is close to the object. Depending upon the platform (e.g. Android, iOS) we've observed focus "hunting" that will impact computer vision algorithm's performance. Thus, small parts/objects may be problematic. Additionally, continuous autofocus when using ARKit & ARCore (highly recommended for the Model Targets feature) can tend to be slow to react which causes additional problems for a user's experience.

Hope this helps with your evaluation.

Thanks,

Vuforia Engine Support

Thank you, but not detailed enough.

How well does Model Target handle small objects such as USB connector or sockets on an electronic device?

See attachment please.

Hello,

Please review the following article and let us know if you have any specific questions:

https://library.vuforia.com/model-targets/model-targets-supported-objects-cad-model-best-practices

Thanks,

Vuforia Engine Support