[benjikan]

Thanks for the info, Ben. Your sharpening technique is excellent and I use it often.

I would like to know more about the application of the method you use and how you do it. Is it done in Photoshop?

The image you posted looks stunning.

Cheers,
Lance

Hello Lance;

All of my post prod is done from ACR and to Photoshop CS2. Tweaking in ACR, major modifications in PS CS2.

Ben

[Matjazz]

I'm well aware of Genuine Fractals but is there a free program that does fractal resizing? I'ev searched the net with no luck. However I stumled upon a site elaborating about upsizing images. They produced results better than Genuine Fractals IMHO.



bicubic:


Genuine Fractals:


Their implementation of fractal based zooming:


Too bad they didn't develop something more than a demonstration program. They have source code available thou but that's no use to me.

Oh, here is the page: Image Resizing - outperform GDI+ - The Code Project - C# Programming

[Matjazz]

I think I found something here: http://meesoft.logicnet.dk/

Download ImageAnalyzer and then scroll down for plugins where you'll find Fractal interpolation.
First impressions are good.

[TomE]

Ben, how about some info on how to do this like the step by step you did for the sharpening information?

Tom

[philmorley]

There are ways to improve the rendition of the K10D that will give it similar resolution to a Pro Back of 31.6 megpaixels. I will be shortly doing an Ad Campaign with the K10D where I could have opted for a Digital Back. I will shoot in RAW and introduce three layers with a microscopic shift on each of the layers. After doing so I will sharpen the layers at between 0.3 and 0.5 pixels at 240 to 300 percent. It works very well. What I do is use two colour layers and one B&W layer. I still use the "Soft Light" setting. I am effectively getting the impression of a 30.6 megapixel image. The pixels are all slightly offset. It really does work quite well.

Thought I would share that with you. Theoretically it is not 31.6 mega pixels, but it sure looks like it.

Ben

hi, thanks very much for this. I really like your stuff (just spent a good hour or so of works time looking through your portfolio - oh well anyway will try it tonight. Is there are reason for sharpening after duplicating? just wondering as my thought would be sharpen then duplicate?

Regards
Phil

[benjikan]

hi, thanks very much for this. I really like your stuff (just spent a good hour or so of works time looking through your portfolio - oh well anyway will try it tonight. Is there are reason for sharpening after duplicating? just wondering as my thought would be sharpen then duplicate?

Regards
Phil

I normally sharpen before retouching. However in this case the re-sharpening does more than just USM it rectifies somewhat the offset that you have done prior to the USM. I suggest trying from 0.3 to 0.6 pixels at 180 to 300 percent per layer...

Dr. B. Kanarek
MD USM non-certified

[TomE]

Ben, which way do you offset each layer, I tried both left and right with the two duplicate layers and it seemed to work out OK?

Tom

[amateur6]

I'd wondered about something like this in the past (I'm an Art Director & Designer who does a lot of advertising and catalog work for Swiss Army) but never played with it. Still, it seems to me like it shouldn't do a better job than any other interpolation...

Process: Duplicate, free transform to reduce the duplicate's scale by 1 pixel (as ennacac said, does the direction matter?), repeat this for another, desaturated layer duplicate of the original... should both of the duplicates be set to "soft light"? What opacity?

I'll also echo kmccanta:

I dont understand. If your microscopic shift is less than a pixel, that means you are creating new pixels interleaved with the originals. Isnt this upsizing?

In other words how can you shift by less than a pixel unless you upsized to define the intermediate pixels to begin with??

I guess I dont understnad how to do this either...

Thank you sir!

[benjikan]

I'd wondered about something like this in the past (I'm an Art Director & Designer who does a lot of advertising and catalog work for Swiss Army) but never played with it. Still, it seems to me like it shouldn't do a better job than any other interpolation...

Process: Duplicate, free transform to reduce the duplicate's scale by 1 pixel (as ennacac said, does the direction matter?), repeat this for another, desaturated layer duplicate of the original... should both of the duplicates be set to "soft light"? What opacity?

I'll also echo kmccanta:



Thank you sir!

Interesting, I am also a Creative Director. Worked on "Dim", "Club Med", "Lancome" etc...I am not saying that it "IS" 30 megapixels, but think of the logic. It is kind of like the "Foveon" sensor, metaphorically speaking. I am basically thickening up the pixels three dimensionally.

[mutley]

Interesting, I am also a Creative Director. Worked on "Dim", "Club Med", "Lancome" etc...I am not saying that it "IS" 30 megapixels, but think of the logic. It is kind of like the "Foveon" sensor, metaphorically speaking. I am basically thickening up the pixels three dimensionally.

OK Ben,

HOW ITS DONE:

Step 1:




















Please let us know

Larry

[benjikan]

If I "Free Transformed" all at the same pixel offset there would be no effect. You take the original for example, I don't know let's say 2400 x 3600 as the original layer. The first duplicate I would make 2399 x 3599 and the second 2401x3601 and when applying Unsharp Mask I would sharpen at 0.3 to 0.5 pixels at 240 to 300 percent to compensate for the offset. Do not Flatten the layers as that would negate the size of the file.

Just remember to sharpen after the fact and on each layer to compensate for the offset. Try between 0.3 & 0.6 pixels at from 180 to 300 percent. I don't know which sensor you are using. I normally sharpen before the retouch and you are doing so as well here. Except your doing after the import from Raw and after the layers.

[mattdm]

I'm not quite sure what you mean by not flattening the image. Obviously it ends up flat.

But anyway, what I wanted to comment on was fractal resampling algorithms. Notice that the example picture is a woodland scene, exactly where this approach is strongest. It's not good for general purpose use, though, since it works by inventing artificial detail, which looks obviously poor if the image in question is a sign with lettering, or details of a human face.

[amateur6]

Interesting, I am also a Creative Director. Worked on "Dim", "Club Med", "Lancome" etc...I am not saying that it "IS" 30 megapixels, but think of the logic. It is kind of like the "Foveon" sensor, metaphorically speaking. I am basically thickening up the pixels three dimensionally.

I tried being a Creative Director, couldn't take the stress. Dropped back to AD and picked up my camera!

Sorry, I absolutely didn't mean to imply that you were making untrue claims! Hmmm... could this be applied at the individual channel level for greater effect? Just thinking aloud.

Inspired, I tried a variety of upsampling techniques and posted them on my now newly inagurated blog, here: http://www.studio610.com/wp/ (if anyone's interested -- I don't have any pretty models, though). Maybe over the weekend I can try another pass at your technique for comparison, Ben. I can also supply my (oh-so-proprietary) test image if anyone has any other workflows they can contribute...

[Mike Cash]

I'm positive I haven't done this exactly as Benjamin outlined, but even in my own muddling way I can see what a difference the method makes.

Here is a before and after comparison of a test I did on a semi-macro shot I took of one of my gladhands this afternoon:

[benjikan]

I'm positive I haven't done this exactly as Benjamin outlined, but even in my own muddling way I can see what a difference the method makes.

Here is a before and after comparison of a test I did on a semi-macro shot I took of one of my gladhands this afternoon:

Here is your image less blown out.....