## Super-Resolution with Convolutional Neuronal Networks, why interpolation at the beginning?

1

I have read several papers about super-resolution with CNNs, where a low-resolution image is reconstructed to a high-resolution image. What I don't understand is, why it is necessary to interpolate the low-resolution image at the beginning to a size that matches the high-resolution image target.

What is the idea about that? If I have an image to image transformation, what are the benefits in a Neuronal network to have the same input size as the same output size?