YOLO算法改进指南【初阶改进篇】:2.改进DIoU-NMS,SIoU-NMS,EIoU-NMS,CIoU-NMS,GIoU-NMS

非极大值抑制(Non-maximum Suppression (NMS))的作用简单说就是模型检测出了很多框,我应该留哪些。
YOLO算法改进指南【初阶改进篇】:2.改进DIoU-NMS,SIoU-NMS,EIoU-NMS,CIoU-NMS,GIoU-NMS_第1张图片
本篇将演示如何修改:NMS、Merge-NMS、Soft-NMS、CIoU-NMS、DIoU-NMS、GIoU-NMS、EIoU-NMS、SIoU-NMS

1. NMS过程

NMS过程
For a prediction bounding box B, the model calculates the predicted probability for each category. Assume the largest predicted probability is p, the category corresponding to this probability is the predicted category of B. We also refer to p as the confidence level of prediction bounding box B. On the same image, we sort the prediction bounding boxes with predicted categories other than background by confidence level from high to low, and obtain the list L. Select the prediction bounding box B1 with highest confidence level from L as a baseline and remove all non-benchmark prediction bounding boxes

你可能感兴趣的:(《YOLO算法改进指南》,YOLO,算法)