DSRL

DSRL copied to clipboard

How to apply this work to BiSeNet

2

Hello, Dootmaan! I want to apply it to BiSeNet, can you give me some advice?

zoe218

Is there a prediction code?

1

Is there a prediction code?

DSQ111

关于上采样

1

您好，非常感谢您的分享，我想请问您， x_seg_up = F.interpolate(x_seg, size=input.size()[2:], mode='bilinear', align_corners=True) x_seg_up = F.interpolate(x_seg_up,size=[2*i for i in input.size()[2:]], mode='bilinear', align_corners=True) 这里为什么要上采样两次，而不是直接一次上采样到原始大小呢？

kaigelee

I can't seem to trained the model to get good result

2

Hi there, Thanks for sharing your codes. I wonder if the experience setting in your current train.py can achieve mIoU=0.6768 on Cityscapes. I trained the model, but I can get...

gnsci

关于SISR分支

1

您好，非常感谢您的分享，我想请问您，本文中提到的超分辨率分支SISR的实现代码是在哪个文件中呢？文中提到的MSE Loss是在哪个文件中呢？

didididj

about input size

2

I see the model's input reduce by half, SSSR and SISR output's size is origin input size, i think the output size is twice bigger than origin input size.

Ontheroad123

About the feature-level visualization

1

Hi, Thanks for your good work. I have a question about Figure3 in the paper: what is the way you do the feature level visualization. Thank you.

manmanjun

dataset instructions

1

Hi. I'm a noob. I'm able to follow your clear instructions so far. (Thanks!) I'm not sure exactly about the datasets I need. I downloaded "checkpoint.pth.tar" and put it in...

SB2020-eye

some questions about environment

3

Thank you for your work! I would like to ask about your GPU configuration and the specific version of some dependent packages used

rubykaguya

about half the amount of fea_seg are zero

2

Hi there, thank you for your works. i have git clone your code, and i have cancel the comment of line 16, 18, 23, 25 in 'util/fa_loss.py' and run 'train.py'...

nickin1023