To train an object detection model to recognize your objects, you have to gather images that contain those objects.
The first thing you'll do when you create an AI Builder object detection model is to define its domain.
Upload images
Then Train and publish your object detection model
The object detector component takes a photo or loads an image file to do an object detection scan. On a mobile device, the user chooses between taking a photo or selecting one already available in the device user interface. When an image is selected, the component automatically scans it to identify objects.
Reference:
https://docs.microsoft.com/en-us/ai-builder/prediction-overview
https://docs.microsoft.com/en-us/ai-builder/object-detector-component-in-powerapps