Around the year 2019 I started to get interested in the concept of AI and trying to understand how it worked. A lot of the early introductory guides at that time were still focused on image classification and object detection models. This is because, image classification specifically, were very easy to implement and well understood at this time.

As I was just getting started in the field. I decided to create an app that would use an AI image classification or object detection model to perform some task in order to teach myself about the technology. I decided to try to recreate the device from the show Pokemon as I grew up watching that, so it seemed like a good place to start. There was also a lot of images to create a good dataset to use.

I ended up integrating my image classification AI into an app. An earlier production release of this app had a real time objection detection model I made, but I removed that as I didn’t like how much space it took up compared to the image classification model.