In this chapter, we talked about numerous object detection concepts, such as HOG, BOW, SVM, and some useful techniques, such as image pyramid, sliding windows, and non-maximum suppression.

We introduced the concept of machine learning and explored the various approaches used to train a custom detector, including how to create or obtain a training dataset and classify data. Finally, we put this knowledge to good use by creating a car detector from scratch and verifying its correct functioning.

All these concepts form the foundation of the next chapter, in which we will utilize object detection and classification techniques in the context of making videos, and learn how to track objects to retain information that can potentially be used for business or application purposes.