FGFA资料

光流介绍

https://blog.csdn.net/u013010889/article/details/71189271

Flownet

https://blog.csdn.net/zjc8888888888/article/details/78171018
https://blog.csdn.net/hysteric314/article/details/50529804

ResNet

简述> https://blog.csdn.net/u013181595/article/details/80990930
论文解读> https://blog.csdn.net/c_chuxin/article/details/82948733

Inception-v4, inception-resnet and the impact of residual connections
on learning.:

https://www.jianshu.com/p/006248a3fd7f
https://www.cnblogs.com/zhhfan/p/10134117.html
https://www.cnblogs.com/eniac1946/p/8708416.html

R-FCN

https://blog.csdn.net/WZZ18191171661/article/details/79481135
https://www.cnblogs.com/lillylin/p/6277094.html
https://www.jianshu.com/p/db1b74770e52

SGD

https://blog.csdn.net/wfei101/article/details/79490470

其他了解

https://www.zhihu.com/question/52185576
https://www.cnblogs.com/wyz2016/p/7420227.html

FGFA代码：https://github.com/msracver/Flow-Guided-Feature-Aggregation

动态图：https://python.ctolib.com/guanfuchen-Flow-Guided-Feature-Aggregation.html

整体了解：https://blog.csdn.net/u012426298/article/details/80487453
简略：

https://blog.csdn.net/zhangjunhit/article/details/76684849

各种滤波：

http://www.mamicode.com/info-detail-2322692.html

卷积核之类的计算：

https://zhuanlan.zhihu.com/p/29119239

Seq-NMS

http://tongtianta.site/paper/850

图像插值

https://blog.csdn.net/qq_37577735/article/details/80041586#1双线性插值

Deformable convolutional networks.

https://www.jianshu.com/p/206e7b0cb433
https://www.cnblogs.com/daihengchen/p/6880774.html

Deep feature flow for video recognition（论文中的检测网络用了这个算法）

https://www.cnblogs.com/hf19950918/p/10588182.html
https://blog.csdn.net/u010213183/article/details/72863829