In the previous chapter, we discussed, in detail, the various neural network image classification and object detection architectures that utilize multiple steps for the object detection, classification, and refinement of a bounding box. In this chapter, we will be introducing two single-stage, fast object detection methods—You Only Look Once (YOLO) and RetinaNet. We will be discussing the architectures of each model and then perform inference in real images and videos using YOLO v3. We will show you how to optimize configuration parameters and train your own custom images using YOLO v3.
The topics covered in this chapter are as follows:
- An overview of YOLO
- An introduction to Darknet for object detection
- Real-time prediction using Darknet and Tiny Darknet
- Comparing YOLOs – YOLO versus YOLO v2 versus YOLO v3
- When to train a model?
- Training...