YOLO

YOLO is a real-time object detection system based on deep learning and is included in the Darknet framework. Its name comes from the acronym You Only Look Once, which references to how fast YOLO can work. On the website (https://pjreddie.com/darknet/yolo/), the author has added an image where this system is compared to others with the same purpose:

Figure 9.1: A comparison of object detection systems

In the preceding graphic, the y axis represents the mAP (mean Average Precision), and the x axis represents the time in milliseconds. So, you can see that YOLO achieves a higher mAP in lesser time than the other systems.

It is also important to understand how YOLO works. It uses a neural network, which is applied to the entire image and splits it into different parts, predicting the bounding boxes. These bounding boxes are similar to rectangles marking off certain objects, which will be identified later in the process. YOLO is fast, because it is able to make predictions with only an evaluation...