Converting a TensorFlow model developed using the TensorFlow Object Detection API

This section describes how to convert a frozen graph created using the TensorFlow Object Detection API. If the model is developed using the TensorFlow Object Detection API, then the detailed process is different to what we covered in the previous section. More information can be found at https://docs.openvinotoolkit.org/latest/_docs_MO_DG_prepare_model_convert_model_tf_specific_Convert_Object_Detection_API_Models.html.

The model that we obtained in Chapter 10, Object Detection Using R-CNN, SSD, and R-FCN, using the TensorFlow Object Detection API is already frozen. Follow these steps to convert it:

1. Before converting, refer to the following link to configure the model optimizer: https://docs.openvinotoolkit.org/latest/_docs_MO_DG_prepare_model_Config_Model_Optimizer.html.

2. Now, we are ready for...