Lynn Twiss

Yea, I cheat. I have a good camera, but lighting isn’t my game. For years I’ve done loads of Photoshop work and it comes pretty naturally to me. However, in the past year I learned some invaluable skills from a master color-correctionist: David Hunt of Turkshead Productions. This man is a dynamo. I’ve never seen human fingers fly across a keyboard the way his do. Even secretaries don’t type as fast as he fires out Photoshop binding key strokes.

While I was in Seattle doing some training with him, I was finally able to see and begin mastering how to do color correction, without the use of Adobe’s Auto Corrects, Levels or even a full color image. Instead, I now do my correction in black and white. I tweak each channel individually. This enables me to pull color data from one channel and mix/blend/add/extract into/out of another channel. Sometimes the yellow (Y) channel has better skin data than the Cyan (C) channel etc.

Here is an example.

This bored morning of mine I took the following picture of myself. (I look so sad!)

Please note how nasty the contrast and white balance is. I shoot in RAW, so pulling that into Photoshop CS2 and tweaking the WB is very easy.

From there I convert it from its native RGB 3-color embedment into a CMYK 4-color spread. I then contrast level all except for the Black (K) channel (contrasting the K level you loose massive amounts of data and clarity). This gives very punch and vibrant, but natural looking tones. This is an equivalent to the Telecine process that motion pictures undergo (this coupled with the 24p frame rate gives motion pictures that ‘edge’ over consumer cameras and short-films). Ending up looking like:

If you notice, the ridge of my nose and my lips are red. This is from the seasonal changes (lips) and my eye glasses (nose). So, in order to kill the redness and bring it more in line with my natural skin color I must dig into the red (M) channel now. The reason for this is, if I were to do a hue change in just those areas, we are pushing the color into a saturation of an unnatural tone. So, instead we will dampen the red, by simply removing it.

Using the Quickmask mode, I pull up a brush and feather it, then brush across my nose and lips, just in the areas that need some subduing. Then I pull up either the ‘curves’ or ‘levels’ (I tend to use curves when I’m doing professional work, and levels when its more personal and doesn’t need to be as precise for a magazine spread or a 22 foot long trade show booth image) and pull the dark space more towards a grey middle ground. This is telling Photoshop that the red shouldn’t be as saturated in that area.

After that I combine it all and do a simple sharpening of the image. Then so some simple ‘touch-up’ like removing blemishes and bug bites! (Canon’s CCDs tend to be very ‘smooth’ – even with the custom sharpening enabled. I like a good, detailed and ‘character filled’ face)

Final image (These images are all low resolution, I tend to work in MUCH higher R):

It is a bit cheating, but it’s a very accurate way of maintaining a full bodied color look, without compromising the original color data. (All the images I’ve taken for Kimber employed this process)

I must be bored.