How to use Neural Network without face tracking

Or how to disable Face Tracking to reduce CPU usage

You can disable Face Tracking manually in any effect so that CPU consumption will be significantly reduced.

In our tests we reached 62% total CPU usage only with background segmentation enabled*.

To disable face recognizer just add " frx_disabled" to "recognizer" features inside config.json.

You should have similar structure in your effect:

...

"recognizer" : [

"background",

"frx_disabled"

]

...

Modified example effect: https://drive.google.com/file/d/1GMSAsQU_j6jCMLkVhGdxA0hK6C99qU30/view?usp=sharing

ℹ️ Keep in mind

Some effects can't work with disabled recognizer (e.g. 3D masks), so don't add frx_disabled option if you need 3D-mask in your effect.

If you don't need face tracking at all, ask your sales manager for a new Banuba SDK build without face tracking so that you won't have to add frx_disabled to all your effects manually.


* - on iPhone 11 with iOS 13.