DoorDetect Training and Export

This repository contains code to train, export and test a tensorflow object detection API model, specifically for the DoorDetect-dataset. Additionally there is code to convert/compile the model to a .blob-file for the OAK-devices.

There are three jupyter notebooks:

1. Train and Create Doordetect Model
2. Object Detection From TF2 Saved Model
3. Create a binary blob for an OAK-device from an object detection API model