Home YOLOv3(You Only Look Once)
Post
Cancel

YOLOv3(You Only Look Once)

What is YOLOv3?

YOLO is similar to CNN. However, there are no flattening layers in YOLOv3. The basic structure of YOLO is:



As you can see in the image above, the goal of YOLO is to generate bounding boxes to detect the position of objects. In the process of YOLO, concatenation occurs, copying the image and joining it before applying preprocessing steps. This concatenation enhances the model’s ability to detect objects. Residual blocks are convolutional layers. One advantage of YOLO is its ability to detect the same object at different sizes, including smaller objects.





Implementation

This post is licensed under CC BY 4.0 by the author.