I have this model trained by us, with MakeML to identify real-life objects. . So I am thinking in order to improve the model we can record the user camera and send back shor