Object Detection model and api interface for the "Object Detect" view in my Glimpse app here: https://github.com/rashanarshad/react_deconv
This repo is currently missing the model due to size limits, but the model can be generated using the scripts and weights provided through this tutorial: https://machinelearningmastery.com/how-to-perform-object-detection-with-yolov3-in-keras/
Check out the self generated FastAPI docs here: https://objdetect.rashanarshad.com/docs
After preparing and saving the model as an h5 file, place it in the app directory and build docker container as normal.