KAIST Develops AI That Reconstructs Cloud-Covered Ground in Satellite Images with Photo-Like Clarity (IMAGE)

The Korea Advanced Institute of Science and Technology (KAIST)

KAIST Develops AI That Reconstructs Cloud-Covered Ground in Satellite Images with Photo-Like Clarity

Caption

Changes in generated satellite images with ground sample distance (GSD). In these examples, the descriptive sentence was kept the same while the GSD was changed from 1 meter to 32 meters. For images with the same number of pixels, a larger GSD shifts the coverage from views of individual facilities to broader land-use patterns. For example, an image 256 pixels wide covers a ground width of about 256 meters at a GSD of 1 meter and about 8.2 kilometers at a GSD of 32 meters. These images were generated under different spatial-scale conditions. They do not show a single real location zoomed in or out step by step, and they are not super-resolution results.

Credit

KAIST

Usage Restrictions

No

License

Licensed content

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.