mlkit VS CoreML-Models

I used ML Kit pos estimation API. I followed New features in ML Kit video and building use cases playlist. I tried to understand googlesamples/mlkit and tensorflow/examples but I barely understand, where the define the processor, and I loos it when it comes to the cameraSource file. eventually I created a runnable application.

To understand the implementation of getBitmap(), please refer to BitmapUtils.java.

Next, we draw an overlay above the camera preview. There is an excellent GraphicOverlay class provided by Android ML Kit demo project. We don't need to reinvent the wheel:

Download the sample code of ML Kit vision.

MLKit is distributed by CocoaPods, no Carthage or SMP support, but at least we ask Google to add such options...

Super far fetched idea on passing 4.2 with iOS specific functionality: there are CoreML models specifically capable of identifying car makes and models (linked in this repo), which could allow you to take/select a photo, and automatically identify/search the car based on the prediction. That being said, it will not take into account nearly as many details as you can manually specify in your app. Nice work!

Mac? Use the built in coreml; https://github.com/likedan/Awesome-CoreML-Models