YOLOv4 PyTorch

What is YOLOv4?

YOLOv4 is a real-time object detection model that was published in the April of 2020. It achieved state-of-the-art performance on the COCO dataset for object detection. By using YOLOv4, you are implementing many of the past research contributions in the YOLO family along with a series of new contributions unique to YOLOv4 including new features: WRC, CSP, CmBN, SAT, Mish activation, Mosaic data augmentation, CmBN, DropBlock regularization, and CIoU loss. In short, with YOLOv4, you're using a better object detection network architecture and new data augmentation techniques.

YOLOv4 Procedure

YOLOv4 breaks the object detection task into two pieces, regression to identify object positioning via bounding boxes and classification to determine the object's class. This is similar to the procedure that was used for YOLOv3 (shown below).

Image courtesy of Ethan Yanjia Li

YOLOv4 Results

YOLOv4 performs exceptionally well with both faster speeds and higher mAP than its predecessor, YOLOv3.

Further Reading

How to Train YOLOv4 on a Custom Dataset: https://blog.roboflow.com/training-yolov4-on-a-custom-dataset/
Breaking Down YOLOv4: https://blog.roboflow.com/a-thorough-breakdown-of-yolov4/