I've got one simple question considering many high-res skins out there at the moment, mainly Wile's skins and FF's skins. That is, how do you get such good face maps? Do you make them from scratch or with reference photos found in google?

Ha! found it. The guy changed his website from the link I bookmarked so I had to google a bit.

http://www.pig-brain.com/tutorials.htm

Theres 2 ways to do face textures (that ive seen)

1 photos cut up and modified

or


2 completely from scratch.

It is a combination of photoskinning and actual drawing. Just like in any college art class, the mantra is know your subject and know your shadowing and highlighting.

Let me add that simply photoskinning does NOT look that good. 100% of the time the shading and highlights do not match as well as the fact that the picture will never line up easily. Ever. You need to know how to draw things by hand and how to manipulate photos WITHOUT distorting them to the point of over-pixelation. Many skinners / modelers prefer to hand draw the entire thing, they like that look better. I personally don't. I want to make it look as real as I possibly can. There is no wrong or right side to that arguement; It's akin to people arguing whether live music or studio recorded music is better (studio FTW :) ). It's all a matter of opinion. In many cases it is actually better to draw a skin by hand than try and manipulate a photo (for one, it's a lot faster). But I do not feel that is the case when attempting the human body. As human we are so in tune with how we look (you see a human up-close everyday in the mirror) that any tiny imperfection will be noticed by the human eye and tarnish the illilusion. I made a few simple mistakes in my first skins and boy o boy everybody noticed them immediately. In my opinion hand-drawn faces, no matter how good the artist, almost always end up looking too "perfect", which is a imperfection in itself because we do not look perfect. Except for PickItUp.

I personally start with VERY hi-res photos that were specifically designed and photographed just for 3D model skinning. After you finally get all the basic peices in place you go to work with the standard artist's brushes and tools and basically draw all over the photo until it appears seamless and all the shading and highlighting is correct. The one thing I found in common with all photos of faces I've used is that they simply did not have strong enough shadowing and highlighting for use on a 3D model. Always have to add that stuff in, and heavily. It's like actors putting on stage makeup.... you always think you put too much on, but you actually need even to add even more. It sounds a lot easier than it is actually.

For reference here is an example (http://www.wiledod.com/temp/222.jpg) of the size and quality of a typical reference photo. BTW you may recognize part of it; I used the eye socket area for the redux version of the allies. Notice the slight more shadowing on the right side of the nose (his right)? It does not match the left side. It's that kind of thing you are constantly correcting.

Here is a photo of one (http://wiledod.com/temp/111.jpg) of the only PSD files I still have left showing just how many parts are spliced together just to make the head area and how I align all the peices. Keep in mind this was BEFORE I did any actual drawing of shadows, highlights, details, color adjustment or cropping of the head skin

Not to be off topic, but may I ask what version of PS that is?

LOL.
Old. Photoshop 6. I actually have Photoshop CS installed; I keep using 6 out of nothing more than habit I guess. I think it's up to CS3 now for the current version

Ya, CS3 just got released. Mucho dineros though.

Grazi, Wile. Could you list all the tools you usually use when doing this? Right now I'm thinking Free Transform (maybe Warp or something as well?) and the healing brush, along with dodging, burning, and color tools.

free transform is used a ton, yes, but not so much for those others. As a matter of fact I doubt I've ever dodged ot burnt anything. More often a LOT of tedious freeform masking, eraser tool at 50% is a MUST, and creating shadows and highlights with the paintbrush in black an white and converting them to overlays........ It's really a lot of little things that are pretty numerous to mention, I'll PM ya.

Funny thing about Photoshop is there is usually about 5 different ways to achieve the exact same result. There is no "correct" way to do something :)

I guess I should also mention that the second pic above, showing all the different layers, doesn't really show ALL the layers; many have been merged already, especially the face area. See how the nostrils are still separate? It was the same for the left eye, right eye, lips, right eyebrow, etc.......... All told it probably was upwards of 50-60 individual layers that made up the final facemap. It takes quite a bit of time to do.

As a matter of fact I doubt I've ever dodged ot burnt anything. More often a LOT of tedious freeform masking, eraser tool at 50% is a MUST, and creating shadows and highlights with the paintbrush in black an white and converting them to overlays........ It's really a lot of little things that are pretty numerous to mention, I'll PM ya.

thats interesting, ive heard of the method but dodging and burning are my best friends

thats interesting, ive heard of the method but dodging and burning are my best friendsThey might be very useful tools, but I simply have never really had much success with them or learned to use them well. I am by no means saying they could not be used; I'm just saying I didn't use them :)