Sponsors

Gold




Friends

UdS Logo
MPII Logo
DAGM Logo



Invited Talk,
Thursday, 5.9., 9:00-10:00, Günter Hotz Lecture Hall

The Noise Clinic

Jean-Michel Morel
ENS Cachan

Slides

Joint work with Antoni Buades, Miguel Colom, Marc Lebrun

All images have noise, but this noise may have undergone many distortions. Can we take any image, say, a scanned old photograph, and denoise it? This requires a good denoising method and an accurate noise estimator, both working for "any" image and "any" noise. I'll discuss both aspects, and particularly how to estimate a signal-dependent and scale-dependent noise, particularly noise that has been distorted by JPEG compression.

A prototype of the noise clinic is currently on line here (username : demo, password: demo).

This facility permits anyone to try on their own images and to browse through the results of the others in the archive. The main conclusion of the first experiments is that the noise estimation is the crucial point to get an efficient "noise clinic".




Imprint-Dataprotection