jxhuang0508/fsdr

FSDR: Frequency Space Domain Randomization for Domain Generalization

fsdr's Issues

Where can I get the appendix of this paper？

hi，where can I get the appendix of this paper？

Training code

Hi @jxhuang0508 @Dayan-Guan, thanks for sharing code for your great work.

However, I can only find the testing code. Could you also provide the source code? Thanks!

Also, could you provide the results of VGG, ResNet101 for both FSRD and IBN-Net. I would like to reproduce them.

And, could you also provide the testing code on BDD and Maphillary?

Thanks for your consideration!

About the ImageNet images

Hi authors,

Thank you for your great work. Could you provide more details about the subset of ImageNet you were used for domain randomization?

Largely lower results on BDD and Maphillary

Hi @jxhuang0508 @Dayan-Guan,

I tested your provided models of both deeplabv2 and FCN, and use the advent code to evaluate the three datasets, I got the same results on cityscapes, but got largely lower results on BDD and Maphillary.

DeeplabV2: B=39.37% M=40.66% C=44.75%

FCN: B=36.40% (reported 41.2) M=38.53% (reported 43.4) C=44.86%

Could you provide the testing code for BDD and Maphillary? As well as also provide the training code?

Thanks!

关于Spectrum Analysis的问题

作者您好，我想请教一下，在FSDR论文中的3.2节您提出将RGB图片的每个通道都进行64次band-filter，得到了H×W×3×64的数据；在这之后是又对这个数据使用了band reject filter是吗？ 还有得到的这个数据再加上batch-size已经达到五维了，这样如何参与模型训练呢？