zhangyhuaee KinD issues

zhangyhuaee / KinD

Kindling the Darkness: a Practical Low-light Image Enhancer

297 stars 59 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Fix for 4D color channels: What happens when loading an image with 4 channels instead of 3...

#16 ChilledFerrum opened 2 years ago
0
There are many halos in the results of the Reflectance Restoration Net training

#15 ubyang closed 3 years ago
0
How to test images without high data

#14 eeerpjw opened 3 years ago
0
evaluate.py generates images different from results of paper

#13 T5eng opened 4 years ago
2
The NIQE score

#12 JiaqianLi opened 4 years ago
0
about PSNR and SSIM

#11 Justones closed 4 years ago
1
The size of the input image

#10 zhangxiaopang88 opened 4 years ago
0
I have some questions when I run evaluate.py

#9 Vampire0716 opened 4 years ago
0
I have some questions when I run evaluate.py?

#8 Vampire0716 closed 4 years ago
0
I wonder how to find NPE, MEF and LIME dataset?

#7 anothersin opened 4 years ago
0
Here are several mistakes on the correspondance in "our485", like the 79.png, 735.png. The order of low and high images in "our485" do not match, like the 79.png and 111.png. But I correct it ,the decom net train result seem to bee worse! How do you solve it ?

#6 changruowang opened 4 years ago
1
你好，感谢分享你的代码。我发现下载的训练数据中低光照图像中并没有对应的正常光照图像，比如79.png。因此按照你程序里的直接读取两个目录下的图像列表并排序得到的图像对并不能保证低光照和强光照图是同一张对应的。我修改数据集中的问题后训练效果反而更差了，请问这是怎么回事？

#5 changruowang closed 4 years ago
0
Please share latest model in google drive!

#4 ANSHUMAN87 closed 4 years ago
1
When I test your pre-trained model, I found some problems

#3 hanxuhfut closed 4 years ago
5
The result of training code is different form the checkpoint you provided

#2 lvxiaoqian opened 4 years ago
1
new one

#1 zhangyhuaee closed 7 years ago
0