Issues
- 0
关于PSNR和SSIM的测量问题
#52 opened by ROTTK - 2
- 24
- 0
question about image rescaling problem
#48 opened by Feynman1999 - 0
about training convergence speed
#51 opened by Feynman1999 - 2
The training is unstable
#35 opened by shengkelong - 0
what is train_IRN_x2_finetune.yml used for?
#50 opened by Feynman1999 - 0
Signal high-frequency information recovery
#49 opened by Jiayu668 - 0
H264 compression
#44 opened by ld-xy - 1
why using gt size 144?
#47 opened by junhyoung-ew - 4
您好,请问该怎么做使test.py得到的fake_H图片与原GT图大小一致?
#46 opened by dmndxld - 1
- 0
train IRN_x4??
#43 opened by ld-xy - 4
Questions about network with jpeg
#42 opened by brighteast - 5
图像大小的问题
#15 opened by Fiona2016Xue - 1
Questions about the models
#41 opened by mrgreen3325 - 6
Training time?
#32 opened by Wenchao-Du - 6
Weird artifacts when testing on Kodak dataset
#13 opened by amnesiack - 0
About Training Details
#38 opened by zhuyeye - 7
- 0
- 1
about test execute time
#36 opened by Young-Zzz - 2
Link of pretrained model is broken
#34 opened by KangHuaLi - 1
Why the PSNR value decreases by several DBs when I introduces H.264 encoding losses?
#33 opened by xiaoxiongli - 2
- 2
Separate downscaling from upscaling
#31 opened by felipeinagaki - 1
如何输入LQ,得到HQ
#29 opened by sym1998 - 10
Dosen't work with LQ mode
#4 opened by ajuanzju - 7
- 9
Problem about loss_ce
#14 opened by MC-E - 2
- 4
- 2
About Gaussian Meaning
#26 opened by delldu - 3
- 5
About hyper-parameters lambda
#19 opened by codyshen0000 - 1
A minor mistake?
#23 opened by plutoyuxie - 7
How to calculate the jacobian
#5 opened by zwb0 - 1
Some questions about the experiment?
#21 opened by toddwyl - 5
Training Details
#7 opened by wenbinTse - 1
About loss_ce?
#22 opened by codyshen0000 - 2
About Distribution Loss
#20 opened by codyshen0000 - 1
Why we need to crop border?
#18 opened by johnnylu305 - 5
about trainning by Y channel with yuv files
#17 opened by flyBird-445 - 5
Where is the implementation for stage 2 with full distribution matching loss `L_distr`
#16 opened by JingyunLiang - 3
model output is the same shape as input...
#12 opened by zhangqizky - 3
8 times downscale
#10 opened by volcanoxs - 2
Is this training process correct?
#9 opened by coolKeen - 2
GPU memory requirement
#8 opened by sunbingfeng - 1
- 2
Why we need the HR reconstruction loss?
#2 opened by a514514772