Zwicker Designs Algorithm to Repair Corrupted Digital Images in One Step

Matthias Zwicker, a professor of computer science with an appointment in UMIACS, recently unveiled a new algorithm that incorporates artificial neural networks to simultaneously apply a wide range of fixes to corrupted digital images.

Digital images are often subject to a range of imperfections such as blurriness, grainy noise, missing pixels and color corruption. Zwicker designed an algorithm that can be “trained” to recognize what an ideal, uncorrupted image should look like. This innovation allows the algorithm to address multiple flaws in a single image on one step.

Zwicker collaborated on the project with computer graphics experts from the University of Bern in Switzerland.

The researchers presented their findings on December 5, 2017 at the 31st Conference on Neural Information Processing Systems in Long Beach, California.

Go here to read the full news release.