Mastering Computer Vision with TensorFlow 2.x

Model conversion and inference using OpenVINO

This section discusses how to use a pre-trained model or a custom trained model to make inference using OpenVINO. Inference is the process of using a model to perform object detection or classification and is divided into three steps:

1. Running inference using a pre-trained model from ncappzoo.
2. Converting the custom model into IR format for inference.
3. Summary of all the steps involved in a flowchart.

These steps will be discussed in detail in the following subsections.