phoenix104104 / LapSRN

Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution (CVPR 2017)
http://vllab.ucmerced.edu/wlai24/LapSRN/
449 stars 120 forks source link

The Set5 (x4) result reported reproduce #9

Open xp739 opened 6 years ago

xp739 commented 6 years ago

Thanks for your code, I can't reproduce the result with my own model when training the T91 BSDS200 and General100 images. The PSNR of Set5(x4) only 29.7. but the papers PSNR of Set5(x4) is 31.54. Can you provide some details about the training parameters ?

ch135 commented 5 years ago

Execuse me!Do you have HR picture of T91,BSDS200 and General100 for training?

elenacliu commented 2 years ago

The author has put the links of training and testing datasets to /datasets/ directory, and according to issue11 they are all HR images. The author generated LR images by himself.

notknowmuch commented 5 months ago

The author has put the links of training and testing datasets to /datasets/ directory, and according to issue11 they are all HR images. The author generated LR images by himself.

Excuse me, do you have access to the dataset now? I find the link became 404 not found.

WechatIMG197

elenacliu commented 5 months ago

sorry I cannot access the data either.