目标检测之YOLO算法:YOLOv1,YOLOv2,YOLOv3,TinyYOLO,YOLOv4,YOLOv5,YOLObile,YOLOF,YOLOX详解(转)

YOLO官网:

YOLO v.s Faster R-CNN:

1.统一网络:YOLO没有显示求取region proposal的过程。Faster R-CNN中尽管RPN与fast rcnn共享卷积层,但是在模型训练过程中,需要反复训练RPN网络和fast rcnn网络.相对于R-CNN系列的"看两眼"(候选框提取与分类),YOLO只需要Look Once.

2. YOLO统一为一个回归问题,而R-CNN将检测结果分为两部分求解:物体类别(分类问题),物体位置即bounding box(回归问题)。

2. YOLOv1: You Only Look Once: Unified, Real-Time Object Detection

目标检测之YOLO v1算法: You Only Look Once: Unified, Real-Time Object Detection:

3. YOLOv2 (YOLO9000: Better, Faster, Stronger)

目标检测之YOLOv2 算法-YOLO9000: Better, Faster, Stronger:

4. YOLOv3: An Incremental Improvement

目标检测之YOLOv3算法: An Incremental Improvement:

5. Tiny YOLOv3

目标检测之Tiny YOLOv3算法:

6. YOLOv4: Optimal Speed and Accuracy of Object Detection

目标检测之YOLOv4算法: Optimal Speed and Accuracy of Object Detection:

7. YOLOv5算法

目标检测之YOLOv5算法:

8. YOLObile算法

YOLObile:面向移动设备的「实时目标检测」算法(AAAI 2021):

9. YOLOF算法

YOLOF:You Only Look One-level Feature(CVPR 2021):

10. YOLOX算法

YOLOX: Exceeding YOLO Series in 2021

增添目标检测数据集PASCAL VOC和COCO详细解析:

  1. 目标检测数据集PASCAL VOC详解:

2. 目标检测数据集MSCOCO详解:

[1][2][3][4][5]

参考

  1. ^V1,V2,V3参考地址: https://blog.csdn.net/App_12062011/article/details/77554288
  2. ^V4转载地址: https://mp.weixin.qq.com/s/Ua3T-DOuzmLWuXfohEiVFw
  3. ^一文读懂YOLO V5 与 YOLO V4 https://zhuanlan.zhihu.com/p/161083602
  4. ^近距离观察YOLOv3 https://zhuanlan.zhihu.com/p/40332004
  5. ^Faster-RCNN的anchor和YOLOv3的anchor一样吗 https://blog.csdn.net/xiqi4145/article/details/86516511
posted on 2022-05-20 17:15  NLazyo  阅读(342)  评论(0编辑  收藏  举报