An overview of YOLO

We learned in Chapter 5, Neural Network Architecture and Models, that each published neural network architecture improves on the preceding one by learning its architecture and features and then developing a whole new classifier to improve the accuracy and detection time. YOLO was at the Computer Vision and Pattern Recognition Conference (CVPR) 2016 by Joseph Redmon, Santosh Divvala, Ross Girshick, and Ali Farhadi in the You Only Look Once: Unified, Real-Time Object Detection paper https://arxiv.org/pdf/1506.02640.pdf. YOLO is an extremely fast neural network that detects multiple classes of objects all at once at the astounding speed of 45 frames per second (base YOLO) to 155 frames per second (fast YOLO). As a comparison, most cell phone cameras capture videos at around 30 frames per second, while high-speed cameras capture videos at around 250 frames per...