search

swz30 / MIRNet

[ECCV 2020] Learning Enriched Features for Real Image Restoration and Enhancement. SOTA results for image denoising, super-resolution, and image enhancement.
Other
667 stars 96 forks source link

Processing for different upscaling factor #6

Closed akineeic closed 3 years ago

akineeic commented 3 years ago

Thanks for your wonderful sharing first! And I have a small question about how you handle different upscaling factors in the model for super resolution task, since I havn't found the description in the implementation details parts of your paper.

adityac8 commented 3 years ago

Hi The RealSR version1 dataset proposed in ICCV 2019 have the same dimensions for LR and HR. Hence, we do not need to explicitly handle the upscaling factor. In case you want to use our model for other datasets(where LR and HR have different resolutions), you could add this module at the end of our network. Thanks

akineeic commented 3 years ago

Thanks for your explanation!