Towards better image denoising with a self-supervised post-correction network. (IMAGE)

GIST (Gwangju Institute of Science and Technology)

Towards better image denoising with a self-supervised post-correction network.

Caption

Researchers from Gwangju Institute of Science and Technology in Korea, VinAI Research in Vietnam, and University of Waterloo in Canada have proposed a new method to improve the quality of path-traced visuals using a post-correction network and a self-supervised machine learning framework. The model can be trained on the fly to output high-quality images in just 12 seconds.

Credit

Bochang Moon from Gwangju Institute of Science and Technology, Korea

Usage Restrictions

Cannot be reused without permission

License

Original 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.