Yolo algorithm is a target detection algorithm. The goal of target detection task is to find all the regions of interest in the image, and determine the location and category probability of these regions. Deep learning methods in the field of target detection are mainly divided into two categories: two-stage target detection algorithm and one-stage target detection algorithm. The two-stage method is that the algorithm generates a series of candidate bounding boxes as samples, and then uses convolutional neural network to classify these samples.

Yolo algorithm principle

Due to the adoption of deep convolutional neural network, it absorbs many excellent ideas of classical convolutional neural network architecture, and achieves the best performance in position detection and object recognition (real-time detection can be achieved even if the accuracy is very high). Because the author also opened the code. Sincerely praise the author's selfless mind.

The fly in the ointment is that although the code is open source, it is very vague when introducing the architecture principle in the paper, especially some important improvements are basically brushed aside. There are many explanations about YOLO principle on the Internet now.