DONG
DONG
> Thank you for attention. > > Our method is an angular-based method. In your pipeline, please make sure the output of fullconnected-layer is the cosine value(In other words, the...
@huangyangyu After I finish the training process with "resnet100 -->fullconnected-layer --> arcface_layer --> noise_tolerant --> softmax" pipeline, the performance of this model is worse than training directly with arcface. 1)...
using r34 as backbone, the weight training with ms1m changes like [this](https://youtu.be/43ZKKQxD_uk), and training with vgg changes like [this](https://youtu.be/g4LtAnSoRBc). There may be some problems.
> @guangdongliang hello,would you like to share your code?cause,i also reproduce the code in mxnet recently,but i'm not sure it works. I am very sorry! If I put the code...
> @guangdongliang What are the margins? @tianzhi0549 The "box AP" and "mask AP" of yours is 39.7 and 35.7 which is bigger than mine(39.58 35.43)
你好！非常感谢您的代码分享，请问该代码支持IOS9以上的系统吗？因为FairPlay需要反编译进行破解，非常困惑。