How can I modify the "Detect and Track face" example in Simulink for acquiring video from an Android device, obtaining the apk file?

I've tried in this way ( attached file), but it doesn't work.

[Sruthi Yenugula]

Hi Elisa,

What issues are you facing with the model?

[Elisa Mangiameli]

Hi,

Thanks for your interest.

In the diagnostic viewer I get some errors like:

"Simulink does not have enough information to determine output sizes for this block. If you think the errors below are inaccurate, try specifying types for the block inputs and/or sizes for the block outputs" and

"Declaration of upper bound size as Inf is not supported. Please consider enabling dynamic memory allocation to allow unbounded sizes".

I've attached the diagnostic file in which there are all the issues with the model.

Answer 1

[Sruthi Yenugula]

Hi Elisa,

Computer vision functions are used in this Simulink model, those functions depend on OpenCv library. Currently those functions cannot be deployed on Android.

"Simulink does not have enough information to determine output sizes for this block. If you think the errors below are inaccurate, try specifying types for the block inputs and/or sizes for the block outputs" and

To fix the above issue, you need to check the dynamic memory allocation for MATLAB functions setting in config set. But even after doing that you will get build error due to the above reason.

[Elisa Mangiameli]

Hi Sruthi,

thanks for your useful answer, now I’ve understood the problem.

Could you tell me if there is another way to obtain a face detection system on Android, using Matlab?

Answer 2

[Sruthi Yenugula]

You can use this example

https://www.mathworks.com/help/supportpkg/android/ref/detect-and-track-face-on-an-android-device.html