researchmm / AOT-GAN-for-Inpainting

[TVCG'2023] AOT-GAN for High-Resolution Image Inpainting (codebase for image inpainting)
https://arxiv.org/abs/2104.01431
Apache License 2.0
403 stars 64 forks source link

Own dataset #15

Open aniol6 opened 3 months ago

aniol6 commented 3 months ago

I get the following error:

images_masked = (images * (1 - masks).float()) + masks RuntimeError: The size of tensor a (256) must match the size of tensor b (455) at non-singleton dimension 3

The images that I use have a size of 1280x720 (images and masks). I use "pconv" as the mask type. Any suggestion?

KonradZBK commented 1 month ago

Solved the problem? I'm also struggling on this one