You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have test it for custom images(include human and dog , low resolution).It, the upsample image, has no difference with the resize image, sometimes even wrong than resize😂.
But for the example image, it does help, so it is not "an off-the-shelf superresolution model" which conflict with the paper ("Figure 3. Our GAN-based upsampler, similar to Figure 2, can also be used as an off-the-shelf superresolution model for real images with a large scaling factor by providing an appropriate description of the image")
In the situation of upsampling image, we does not have super resolution image (if we have, we do not need it).
The text was updated successfully, but these errors were encountered:
I have test it for custom images(include human and dog , low resolution).It, the upsample image, has no difference with the resize image, sometimes even wrong than resize😂.
But for the example image, it does help, so it is not "an off-the-shelf superresolution model" which conflict with the paper ("Figure 3. Our GAN-based upsampler, similar to Figure 2, can also be used as an off-the-shelf superresolution model for real images with a large scaling factor by providing an appropriate description of the image")
In the situation of upsampling image, we does not have super resolution image (if we have, we do not need it).
The text was updated successfully, but these errors were encountered: