DIV2K is a popular single image super-resolution dataset that can be used to reconstruct high-resolution images from low-resolution images. This dataset contains 1000 low-resolution images with different degradation types, divided into:

• Training data: 800 low-resolution images, and both high-resolution and low-resolution images for the degradation factor.
• Validation data: 100 high-resolution images for generating low-resolution images.
• Test data: 100 diverse images, used to generate low-resolution images.