You are viewing the course site for a past offering of this course. The current offering may be found here.
Lecture 22: Image Processing (25)
nathanpetreaca

Currently, there are deep learning algorithms that are used for denoising and they work pretty well, here is an example https://arxiv.org/pdf/1608.03981.pdf .

gprechter

This reminds me of a paper I've read about image processing recently. It doesn't necessarily have to do with noise, but the above comment reminded me of this paper. It's on Image Super Resolution: https://ieeexplore.ieee.org/document/7115171 Essentially, the idea is to correctly increase the resolution of an image. This method implies deep learning, and while a naive attempt may leave an image filled with noise, by incorporating deep learning, the researchers were able to create an algorithm that could upscale images quite well.

tyleryath

I find it very interesting that some deep learning techniques approach image de-noising by training their models on noisy images, and not the other way around.

You must be enrolled in the course to comment