search

VITA-Group / EnlightenGAN

[IEEE TIP] "EnlightenGAN: Deep Light Enhancement without Paired Supervision" by Yifan Jiang, Xinyu Gong, Ding Liu, Yu Cheng, Chen Fang, Xiaohui Shen, Jianchao Yang, Pan Zhou, Zhangyang Wang
Other
918 stars 200 forks source link

exchange trainA and trainB #122

Closed aimen123 closed 1 year ago

aimen123 commented 2 years ago

I want to exchange trainA and trainB, can we achieve the effect of darkening? Can the code be completed? thanks!

yifanjiang19 commented 2 years ago

I dont try this before but you can definitely try and see hot it performs, by just exchanging the dataset

aimen123 commented 2 years ago

I try it and fail. It dont work out very well

yifanjiang19 commented 2 years ago

maybe you could upload some results here

aimen123 commented 2 years ago

loss does not converge，so....

yifanjiang19 commented 2 years ago

I think loss will never converge, since it is an adversarial training. Usually I only look at the visual results

wzr1201 commented 6 months ago

hi,have you successfully achieved the effect of darkening?