This chapter gives an overview of object detection and several challenges in modeling a good detector. While there are many methods for detection using deep learning, common categories are one-stage and two-stage detectors. Each of the detectors has its own advantages, such as one-stage detectors are good for real-time applications while two-stage detectors are good for high accuracy output. The difference in accuracy between the models is shown using example figures. We can now understand the choice of object detector and run a pre-trained model using TensorFlow. The various output samples for each show the effectiveness of models in complex images.
In the next chapter, we will learn more about the image recognition problems of segmentation as well as tracking using deep learning methods.