What's new
Photoshop Gurus Forum

Welcome to Photoshop Gurus forum. Register a free account today to become a member! It's completely free. Once signed in, you'll enjoy an ad-free experience and be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!

Chroming/Bronzing


It's difficult. The reason for this difficulty is chrome is highly reflective. It's one of the attributes that makes chrome, chrome.

The basic effect is started by desaturating your picture. Then you smooth out the features with diffusion and median. Once you have done this you play with curves to make the contrast more metallic.

Then the really hard part, you make a displacement map and use it on another picture. You then copy the displaced picture and paste it over your manipulated portrait. You play with masking and the blend modes until it looks right.

This tutorial is a good one for displacement maps. Even though it was written for line art, the technique works for photos and pictures too.
http://www.polykarbon.com/tutorials/displacement/displacement.htm

Stroker is the true displacement master par excellence. He has studied and written more about this filter than has anyone else besides the developer of the filter himself, Kai. If you get stuck just send him a message. But please be warned that he is quite busy and he hates being hit up for info by people who don't make an effort first. So have a picture ready. And nothing half-*ssed either. Trust me. You don't want to be assimilated. 3[
 
If you ever want to get Strokers PDF on using the displacement filter (it's the most thorough info available anywhere) I'm hosting it for download, with his permission, of course. The file is Displace.pdf

http://homepage.mac.com/wellesgoodrich/FileSharing2.html

Even if the Displace filter is too obscure for your needs at this time, I'd suggest getting the file anyway and saving it for future studies.
 
Wow. Can't believe I almost missed this thread.
If you can believe it, I have some new ideas. (Damn you, Moth.)
Hopefully by tomorrow I'll have some good stuff.
So, keep yer panties on.
 
f you can believe it, I have some new ideas. (Damn you, Moth.)

Did it to you again, didn't I? :P

I love to hear your ideas. Any chance it involves a revisit to shapeshifter? %}
 
Kinda, Moth. Yes, but not really. Um...

As it turns out, my newer ideas aren't really that great. Keep running into the same problem of needing a height map. Without a proper height map, it's just fakery and no better than other variations of fakery.

Did you catch my little ramble about vector flow fields and Emboss? That is the direction that I was going in, which is just a slightly different approach to ShapeShifter and Reflection Maps (which rely on a proper height map).

:sigh:

There is no white whale, but I'll probably resume the chase in due time.
 
As it turns out, my newer ideas aren't really that great. Keep running into the same problem of needing a height map. Without a proper height map, it's just fakery and no better than other variations of fakery.

A shame. I know from bitter experience that not every line of thought pans out. Still oft times, what looks like a failure, is really a success misapplied.

Did you catch my little ramble about vector flow fields and Emboss? That is the direction that I was going in, which is just a slightly different approach to ShapeShifter and Reflection Maps (which rely on a proper height map).

I missed that one. Do you have the link? I'd like to see it.
 
Hmm... got moved to The Sink.

Vector Flow Fields and Displace:
http://www.ozoneasylum.com/25242

Rather short ramble for the material that I've collected.
I was looking into anistropic, which led to gradients, which led to edge detection, which led to Canny, which led to Sobel, which led to vector fields.
Toss in some calculus (LaPlace) and you've got a big pot of stew.

Did you know that some Custom kernels are calculus short-cuts?
Even High Pass can be thought of in terms of calc.
If you really wanted to, you could use Gauss for edge detection.
Neat stuff.
 
Neat! :righton:

Did you know that some Custom kernels are calculus short-cuts?
Even High Pass can be thought of in terms of calc.
If you really wanted to, you could use Gauss for edge detection.

I did not know that. Would you care to elaborate? Calculus is the only college class I flunked twice. The trigonmetry kept tripping me up. So you may have to dumb it down for me a little. ;\
 
This guys explains if fairly well:

http://www.pages.drexel.edu/~weg22/edge.html

Notice that his LaPlace kernel is 5x5. I find that a bit extreme and usually use 3x3 with 8 in the center. Either way, make sure Offset is 128.

Also notice that the LaPlace kernel beers a striking resemblence to High Pass (the only real difference is that the LaPlace kernel is linear while High Pass is Gaussian).

You don't really need to know the nitty-gritty of calc as long as you understand the basic ideas of delta and delta change. While I like calculas, I don't like sigma notation and I *hate* integrating.

Few interesting things about the Sobel kernels, but will have to wait.
 

Back
Top