Anyone interested in seeing what it takes to re-create this effect in Gimp?

Definitely, that would be a hard effect to create as with the masking, and using map object to get the text to go along the body, may be easier with the cage transformation tool, but this would be pretty cool to see how it's done.

I actually tried it with 2 kittens hugging each other.One black one white.I used the white one as the text layer and tried a displacement map...didn't work, at least for me anyways.Tried the iWarp tool, and got better results along the edges and what not.

Original image:


This in my opinion would be a real hard challenge for even an experienced Gimp user. =)
I would love to see this get a tutorial by the second issue of The Gimper so i can add it.

Did the first issue ever come out? I never heard back about helping and was never given the access on gimper.net

Did you try using Warp (not i-warp), like in the Candy tutorial on GIMP Users? Use a bump mapped image of a solid object, then make the printed text layer the target? Just a thought, I haven't tried it in this capacity, yet.

I thought David gave you access already Greg. You were already a member so i thought he mentioned giving you access to the Gimper Magazine Staff area of the site.Log into Gimper.net and see if your avatar has Gimper Staff over it.If it doesn't let me know, or David know.

No the first issue is still in the works.

I will give the warp idea a try, sounds promising....thanks.

I got some interesting results with Filters/Distort/Lens Distortion.
Not quite correct, but you can see where the outer edges seem to roll out of the text.


I included the xcf which includes
The original image
White-kitty render
Black-kitty render
lorem ipsum text layer
white bg layer
kitty text overlay layer

Ready to manipulate for anyone who is interested in giving it a go.

I gotta try this sometime, won't be able to try it till the weekend, cause of work. (on break right now)

A combination of Filters / Distorts / Lens Distortion and Filters / Light & Shadow / Apply Lens seems to be a start. Tried a few other effects but not caring for the results as of yet ...

It isn't an easy effect to re-create. =)
I will give it another go this weekend also.

Still fighting with Best Buy to get my brand new power computer (that I have been without for weeks) replaced/repaired. If/When I finally get a fully working machine going, I will be trying this out. I did have luck with this style of art years ago (done with Gimp) but don't remember how I did it.

Sounds good PhotoMaster.Hope you get your baby back soon! =)

PM,
What gives? You're sick, your computer is sick...

Gave it a try using Imageskil's Displacement filter (it's free. Works great in GIMP; not so well in CS5; lol).

http://www.imageskill.com/products_index.html

This is definitely worth a tutorial.

Thanks Lyle ill give it a go with the imageskill filter too. =)

i have no idea how this was done, if we found how to maybe we will get even closer



Alas...this was done manually 20h of work

I ran across this program (not free though but they do have a trial period) that does this PC. I'm nearly 100% sure that's the program that these renders were done in. Still lots of work.

http://www.pingotec.com/

It might seem like I am being over-the-top (I'm not, really - more like overly optimistic), but something tells me that creating images with text is as simple as finding a way to wrap one layer (the text) over a 3D-like normal map. It would involve not only bumping, but warping those points to elevated targets on a mapped image. So, this got me to thinking about G'MIC 3D extrusion and the ability to use dots or wireframes on the image. What if, instead of dots or wireframes, we were allowed to map a layer of choice, instead, namely the text layer of brightly colored, varying sizes of text?

Currently, with a layer of transparent text, the input layer option isn't playing nice with the extruded image. Certainly someone with far superior coding skills could simplify that process by including the ability to permit an input layer to follow the same mapping parameters of dots or wire frames. Am I making sense to anyone?

I tried to play this a hundred different ways, using nothing but GIMP and/or G'MIC (no other outside plug-ins/filters were incorporated). And like others, I'm still not convinced or satisfied with the normal displace, bump, or normal mapping process. It just leaves images too distorted and fuzzy. Maybe Mike is right, cage deformation might be the tool that gets us close.