Slashdot Mirror

Run the free GREYCstoration algorithm on both images, subtract results from original, and pick the one most similar to the original: http://www.greyc.ensicaen.fr/~dtschump/greycstoration/

I have played around with it and wasn't too impressed.
This one looks bad...
http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_parrot.png

This one you can mouse-over and have the image change on you. Again, not that impressed.
http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_claudia16.html

Looks very blurry.

This one is actually okay, but its because its an owl. I'm sure all the blurring is there, its just hard to notice.
The eye's pupil isn't as rounded as it could have been.

http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_chouette2.html

This software is okay at denoising camera photos for free, but I like neatimage and don't mind paying for it.

I have played around with it and wasn't too impressed.
This one looks bad...
http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_parrot.png

This one you can mouse-over and have the image change on you. Again, not that impressed.
http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_claudia16.html

Looks very blurry.

This one is actually okay, but its because its an owl. I'm sure all the blurring is there, its just hard to notice.
The eye's pupil isn't as rounded as it could have been.

http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_chouette2.html

This software is okay at denoising camera photos for free, but I like neatimage and don't mind paying for it.

I have played around with it and wasn't too impressed.
This one looks bad...
http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_parrot.png

This one you can mouse-over and have the image change on you. Again, not that impressed.
http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_claudia16.html

Looks very blurry.

This one is actually okay, but its because its an owl. I'm sure all the blurring is there, its just hard to notice.
The eye's pupil isn't as rounded as it could have been.

http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_chouette2.html

This software is okay at denoising camera photos for free, but I like neatimage and don't mind paying for it.

And if you dont have any pictures database, there's always GREYCstoration:
http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/index.html
It's pretty impressive:
http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/demonstration.html
and works with the gimp.

And if you dont have any pictures database, there's always GREYCstoration:
http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/index.html
It's pretty impressive:
http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/demonstration.html
and works with the gimp.

Wow, I hadn't seen that one. It's definitely real. You can make out the artifacts a lot more visibly in this one. It wouldn't be tolerable for serious photography, but for something like newspaper imagery this would be (is) killer.

I understand, but take a look at some of the demo images that focus on removing artifacts from jpegs, especially the last one of the purple flower. The before image pretty much looks like a very low resolution, heavily pixelated image much like what you would see in old games. Although there is a greater deal of edge definition in jpegs and the pixelation occurs mostly in the gradients, still I wonder if some temporal considerations might help a little more in determining edges, at least in dynamic objects. Might be especially helpful in psx games where the edges are defined fairly well by the model structure itself while the face textures look terrible. I'm just wondering if this algorithm could do a better job than say the superEagle or similar filters traditionally used. Maybe I'll run a few screenshots through the program when I get home today. Another consideration is XVid video compression. It does a terrible job with dark areas of an image. Maybe one could apply this algo there.

http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_subtitle.png

"Sounds like somebody's got a case of the Mondays"

Who said french weren't geeky ?

I'm more impressed, if it's real, of the removal of the cage from the caged parrot photo: http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_zoobird.png

Similar, but less so: http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_parrot.png

I'm more impressed, if it's real, of the removal of the cage from the caged parrot photo: http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_zoobird.png

Similar, but less so: http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_parrot.png

If the inpainting algorithm is incomplete, I am utterly impressed.

http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_poupoune.png

This result is absolutely amazing for a computer algorithm. There is only one spot (below left wing) that I can see any kind of artifact.

Some of the other results are less stupefying, but still impressive. Comparing the time it would take to get the same (most likely worse, if your as bad of an artist as I am) result in GIMP (using the stamp tool mainly), this tool is very impressive. I recently removed text from a picture (the advent children poster), but it took me at least an hour. I bet this tool could handle the same problem in a matter of minutes (creating the mask, and running the algorithm).

Even the other photos which weren't quite as perfect:

http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_claudia.png

could be fixed much quicker by running the algorithm, then going back by hand.

Unfortunately I can't think of too many practicle purposes other than the one I already mentioned.

If the inpainting algorithm is incomplete, I am utterly impressed.

http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_poupoune.png

This result is absolutely amazing for a computer algorithm. There is only one spot (below left wing) that I can see any kind of artifact.

Some of the other results are less stupefying, but still impressive. Comparing the time it would take to get the same (most likely worse, if your as bad of an artist as I am) result in GIMP (using the stamp tool mainly), this tool is very impressive. I recently removed text from a picture (the advent children poster), but it took me at least an hour. I bet this tool could handle the same problem in a matter of minutes (creating the mask, and running the algorithm).

Even the other photos which weren't quite as perfect:

http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_claudia.png

could be fixed much quicker by running the algorithm, then going back by hand.

Unfortunately I can't think of too many practicle purposes other than the one I already mentioned.

They should also use real, not artificial, noise.

Check out very impressive clean up of a PDA camera. That's good. Ordinary smoothing filters blur important details, like those in the watch or the baby's ear. How nice that it is already in Digikam, one of the easiest to use photo managers out there.

They are overly smoothed and detail is destroyed. They look like the type of thing a noob makes upon discovering video filters. For example, look at the delicate features in the jellyfish or the pig's hair. This samples look more like demonstrations of soften or posterization filters.

Sure, he's a noob. That DT-MRI of gray matter paths in your brain based on diffusion tensors is purely the stuff of rank amateurs! Bah, next you will tell me free software authors can make a powerful and easy to use image editor. I'm sticking to well known commercial filters found in Paint Shop Pro version 1.0, you know the one that fits on a floppy. Yeah, that's the freedom program.

Oh wait, those other filters are not helping my pig. Perhaps I'm doing something wrong and need to leave that to the pros as well.

They should also use real, not artificial, noise.

Sufficiently advanced noise is indistinguishable from the stuff that comes out of a cheap imaging device, but it's not magic.

Their examples even include an image of Boba Fett skiing!
http://www.greyc.ensicaen.fr/~dtschump/greycstorat ion/img/res_ski.png

On a more serious note, this look pretty useful. I've been able to get quite good results on many images with a combination of blurring, denoising, unsharp masking and other algorithms in Photoshop and the GIMP, but nothing beats a proper anisotropic diffusion. And in the versions of the aforementioned software I've used, there's no such thing available (maybe in newer versions, so don't shoot me).
Too bad the inpainting part appears unfinished (at least the documentation says "to do"). Inpainting is the reconstruction of missing parts of the image from surrounding parts. I've managed to do successful inpainting by using a simple diffusion kernel (GIMP and PS allow to specify a custom convolution kernel), but this takes ages because you need to repeat the transformation until the image stabilizes, and you can only use dumb diffusion kernels like a square with a zero in the middle.

It seems to perform a lot better when dealing with artificially-added noises than real-life images, as if it already knows how to tackle them.

This pyramid photo has basically been 'ruined' after the denoising, I wonder if we added some synthetic noises in the background while leaving the stone face as is, would this app be able to denoise correctly?

GREYCstoration. Ugly name, but does the same job, and is open source. Haven't tried it, but there appear to be several plugins for various open source digicam programs and image editors (bottom of their downloads page).

GREYCstoration. Ugly name, but does the same job, and is open source. Haven't tried it, but there appear to be several plugins for various open source digicam programs and image editors (bottom of their downloads page).

locally smooth the image preferably along the image structures