How to optimize the existing approach
As mentioned earlier, in order to optimize the existing approach, I will be using the TensorFlow Object Detection API. You can refer to Google's TensorFlow GitHub repo for this API at the following link: https://github.com/tensorflow/models/tree/master/research/object_detection. This API is trained using the COCO dataset as well as the PASCAL VOC dataset; so, it will have the capability of identifying the variety of classes.
Understanding the process for optimization
The most important part for us is how to use the various pre-trained models. The steps are as follows:
First, pull the TensorFlow models repository using this link: https://github.com/tensorflow/models
Once you pull the repository, you can find the iPython Notebook that I have referred to in order to understand how to use the pre-trained model and to find the link for the iPython notebook at https://github.com/tensorflow/models/blob/master/research/object_detection/object_detection_tutorial...
