chainer-partial_convolution_image_inpainting how to test on a real broken image?

how to test on a real broken image?

Open xuanzhangyang opened this issue 6 years ago • 3 comments

Hi, If i have a real broken image, how to test on your model? That means I don’t have the mask as input.

Jun 06 '18 08:06 xuanzhangyang

Hi @xuanzhangyang,
Thank you for your attention.

In this implementation, my model interprets value zero as a broken part (mask) after pixel-wise multiplication between an input image and mask.

So in your case, you can generate the input mask of the broken image by extracting the part of broken pixels and set those value with zero.

Best,

Jun 06 '18 08:06 SeitaroShinagawa

So I need to pick out the broken part of the picture， but sometimes it's hard to do

Jun 07 '18 06:06 xuanzhangyang

Sorry, I misunderstood your question.
I answered to your question in #14.

Best,

Jun 26 '18 16:06 SeitaroShinagawa