Frequency separation is the process of separating the frequencies of light waves in a photograph so that they can be seen as separate images. This is done by using a filter, such as a graduated filter, to reduce the amount of light that reaches the camera at any one time.
What Is Frequency Separation?
Frequency separation works by separating an image’s color and texture into separate layers. This lets you even out skin tones without removing skin texture, and get rid of blemishes without affecting the skin’s color. It’s been used for years and is still one of the most common ways to retouch skin in Photoshop.
This technique can really elevate your photography, but it can also make it look pretty bad if you go overboard. The key to frequency separation, as with any other editing technique, is small, subtle changes that add up to a natural-looking edit. The image below is after frequency separation. A few blemishes have been removed and the skin tones evened out, but the skin texture is largely left alone.
You can see the image before frequency separation below.
There are several different approaches to frequency separation, and no one way is any better than another. It all comes down to what’s easiest for you to do and remember. Once you’ve got the hang of it, there are even Photoshop actions you can download that separate the layers for you — all you have to do is make the edits.
How Does Frequency Separation Work?
The process starts by making two copies of the background layer. Those duplicates will be your high frequency and low frequency layers. Some people name them “color” and “texture” instead since that’s what’s being edited on each layer.
The high frequency layer is where you’ll edit the fine texture and details of an image like blemishes or garment tags. The low frequency layer is where you’ll edit the color and tones of your image, particularly the transitions between light and shadow.
First, you’ll create the low frequency layer, where you’ll work on the skin tones. To remove the texture from this layer, you can apply a gaussian blur effect to the image. With your high frequency layer turned off, go to Filter > Blur > Gaussian Blur. Adjust the radius of the blur until you don’t see any more texture detail in the skin, then turn your high back on by clicking the eye icon next to the name.
Next, you’ll create the high frequency layer, which subtracts everything but the details from the layer below it. This takes a few more steps than creating the low frequency layer. Select the high frequency layer and make sure it’s visible by checking the eye icon next to the layer name. Then go to the top menu in Photoshop and click image > apply image. A dialog box will pop up that looks like this:
In that box, select “low frequency” under the Layer dropdown menu and change blending mode to “subtract.” Once that’s done, hit OK, select the high frequency layer (which will look gray) and change its blend mode to “linear light.” The image should now appear normal and be ready for editing.
This layer is where you’ll edit out things like blemishes, scratches, and other things you don’t want in the image. Photoshop gives you multiple tools to play with here, from the clone stamp tool to the patch tool to the classic spot healing brush.
There are myriad tutorial videos you can watch for the step-by-step process of frequency separation. We recommend watching and trying multiple approaches to give you a better understanding of the “why” behind this technique.
Boudoir and beauty photographer Anita Sadowska offers a great video tutorial on using frequency separation. Another tutorial by YouTube channel Fez Photography demonstrates the lasso tool technique for smoothing skin tones. PixImperfect’s tutorial on using the mixer brush frequency separation technique does a very good job of explaining the science behind frequency separation.
When Should You Use Frequency Separation?
Frequency separation should be used fairly early in the editing process. You can do it after you’ve made Lightroom edits and imported an image, or just after basic global edits like exposure and contrast if you’re only working in Photoshop.
It’s best to leave major stylistic changes like dodging and burning until after this step since frequency separation can change these edits if applied afterward. You wouldn’t want to blend the tones after you’ve taken the time to shape the face with a targeted dodge and burn, for example.
One way to visualize this is to think of the editing process as building a house. Just as it’s difficult to make changes to the foundation after you’ve started building the walls and wiring, it’s hard to go back and change early edits once they’re finished. If you do choose to change them, all the layers after that base layer also have to be changed to match.
Remember, subtlety is key to this technique. Practice with it using reference images until you get an eye for what a truly natural edit looks like. With enough practice, you’ll start to notice your shots have a much more professional look.
