Posing The ModelIn this case, I wanted a strong and confrontational pose and with the hips to the side (normally done to reduce the width across the pelvis), the photo loses a lot of its might. So I just skipped over that guideline and went for the punch rather than concerns about child bearing hip width. Sure, some models might shed a tear or two, but Jane here can handle it.
A term you should get to know is "canoeing", where the bottom white of the eye is showing (makes a canoe shape connecting both sides of the sclera). Originally I didn't much care for this image, as the eyes bothered me for the aforementioned reason, and it only recently dawned on me that I could just paint over the eyes and rescue this photo from the unloved lifetime of my messed-up photo archive.
Lighting the Photo
|thrilling lighting diagram|
The "glow" evident on her far shoulder is from the glare off of the white wall. No reflector was used here as it would have bounced to much light back onto her side, and I wanted the drama. Note that this softbox is gridded to prevent a ton of light from hitting the wall. Controlling the spill from light modifiers is something I find myself doing on every image, as I don't want the background to distract from my subject. If I want to light the background, I will typically do this with another strobe or two rather than use the spill from the key light.
Post Production In PhotoshopMuch of the creepiness of this image is derived from the modified eyes. To create this effect I simply used the pen tool and outlined her eyes, and then filled that selection with black. I then took a white hard-ish brush and painted in where the catchlights ought to be. I just used what was in her real eye as a point of departure here. Also, if this effect is to sharp you will find that using a gaussian blur on that layer will help with the believability. I kept the catchlights on a different layer from the black sclera until I had the eyes the way I wanted them. I then merged them down to a single layer so I could work with the whole image.
I did end up using Liquify a bit to square her hips a slight amount, but otherwise she is pretty much untouched. Once I had her shape finalized I used the quick selection tool to separate her from the background so I could add a texture and the glyph. I played with the blending mode and transparency until I was happy with the result. There are two textures applied here as well as the glyph, for a total of three additional layers. Each has a different blending mode to augment what the previous layer brings to the table.
To help with the punch of the image, I created a 50% gray layer set to the Overlay blending mode. On this I used a 2% white soft brush to add more highlights to parts of her body. This increased the drama of the image, and I then used a black brush to do the opposite in the shadows. Some of you may recognize this as the old method of doing a Dodge & Burn, as the original image layer is unmodified and I can just paint in 50% gray to fix any mistakes. I do this step on all of my images to some degree. This method was needed prior to Photoshop CS3, when the dodge and burn tools in Photoshop should have been named "make more brown", as they sucked at the only job they had. I still prefer the Overlay method because it is non-destructive and quite simple to fix mistakes, not that I make any of course. :-)
One final touch to this image was a solid color "light blue-green" adjustment layer added over the entire photo. I reduced this to around 2%, and it helps bring the blackest parts of the image into gamut and adds a "softness" to the feeling of the shot. I find this effect pleasing and use it on a lot of my low-key images, and it amazes me what this subtle little step can do for the work. This also helps a ton when you get work printed, as the blackest parts of the image are no longer pure black. Another fun alternative to this is to use images of "light leaks" from older cameras. These are all over the web and make some interesting effects vs the solid color I used here.
Any questions on things I might have missed? Thanks a ton of the feedback on the last image gang, was nice to know people are listening!