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

Results of translating SAR observations into optical images. From left to right: the SAR input, the results from the existing methods cBBDM and C-DiffSET, the result from GeoCore-9B, and the actual optical reference image. The pretrained GeoCore-9B was fine-tuned on SAR-to-optical translation data, and its results were then compared with those of the other methods on test images. The figure shows how GeoCore-9B can help users interpret radar data by presenting it with the colors and textures familiar from optical images.

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.