jiangsutx / SRN-Deblur

Repository for Scale-recurrent Network for Deep Image Deblurring
http://www.xtao.website/projects/srndeblur/srndeblur_cvpr18.pdf
MIT License
709 stars 184 forks source link

Cannot feed value of shape (1, 720, 1280, 4) for Tensor 'Placeholder:0', which has shape '(1, 720, 1280, 3)' #55

Closed Lie-huo closed 3 years ago

Lie-huo commented 4 years ago

hello,when i test a 532*570 image, it happens that “Cannot feed value of shape (1, 720, 1280, 4) for Tensor 'Placeholder:0', which has shape '(1, 720, 1280, 3)'”,what‘s the matter?How can solve this problem?

Williamlizl commented 3 years ago

multiple by 16 and set argument when train

jiangsutx commented 3 years ago

And maybe your image has 4 channels, try to convert to RGB images first: you can slice first 3 channels or use opencv