参考文章:
Detectron2基础模型解析:
RetinaNet(Focal Loss):
https://zhuanlan.zhihu.com/p/59910080
一文读懂Faster RCNN
https://zhuanlan.zhihu.com/p/31426458
Detectron2简介和源码:
GitHub榜首:最强目标检测平台Detectron2 ,基于PyTorch完全重构:
https://cloud.tencent.com/developer/article/1521649
Github 项目 - detectron2 安装与简单使用:
https://www.aiuai.cn/aifarm1288.html
Detectron2源码地址:(最新的目标检测技术实现,还是从Detectron2入门比较好)
https://github.com/facebookresearch/detectron2
Detectron源码地址:
https://github.com/facebookresearch/Detectron
Detectron.pytorch源码地址:
https://github.com/roytseng-tw/Detectron.pytorch
Faster RCNN pytorch源码实现,FAIR开发Detectron的起因:
https://github.com/ruotianluo/pytorch-faster-rcnn
https://github.com/longcw/faster_rcnn_pytorch
Detectron2入门教程:
https://blog.csdn.net/weixin_36670529/article/details/104021823
Detectron2内Projects的基础论文解析:
Panoptic Segmentation 全景分割:
https://blog.csdn.net/zziahgf/article/details/79063398
DensePose论文笔记(实现实时的人体姿态估计):
https://blog.csdn.net/Julia_deeplearning/article/details/83011798
一文带你读懂Cascade R-CNN,一个使你的检测更加准确的网络:
https://blog.csdn.net/Chunfengyanyulove/article/details/86414810
Fast Visual Object Tracking with Rotated Bounding Boxes SiamMask_E目标跟踪算法论文笔记: