(Last updated on July 15th, 2021)
Have you ever heard of frequency separation? Do you know how to use Photoshop to get the most out of your photos? If the answer to both questions is no, then read on to learn all about this fascinating aspect of the world’s most powerful creative application.
If you take many portrait photographs of people, you may find yourself trying to tweak how the face looks. You may want to change up the pores or fix some blemishes. However, doing so can be tricky with the tools provided. You may just want to adjust the pores but not mess with the colors.
The solution to this problem is to use frequency separation. It’s a technique that’s become quite common among photographers who want to make minor and major adjustments to their photographs. The good news is that it’s not too difficult to learn either.
Frequency Separation: What Is It Anyway?
Photographers have been using frequency separation in Photoshop to add more information to their photos. It is most commonly used by portrait photographers who want to get their pictures looking crisp and intricate with great detail in their layers.
There are two types of frequencies that generate photo information. High-frequency information applies to the likes of pores, hair, and anything with texture. Low-frequency information applies to the likes of shadows and colors.
If these two frequencies are separate, it becomes easier to edit these specific aspects without affecting one more than the other. The process essentially works with layers and relates to how you coordinate what you want to edit and what you don’t.
Say you wanted to change the look of the skin texture, but you don’t want to affect the same color of the skin. If you separate the frequencies, this can be accomplished. Essentially, frequency separation grants you greater control over fine-tuning details of your photographs in skin and textures.
Creating Frequency Separation Inside Photoshop
Creating a frequency separation starts as simply as creating two layers. With your photo inside Photoshop, make two duplicate layers, so you have two layers of the same photo, with the original as a safety. Give them distinct names such as “High Frequency” and “Low Frequency” to tell them apart, with the high-frequency layer being on the top.
For better organization, consider placing both of these layers into their Group Layer folder. This will not only keep everything organized but prevent confusion when working with other layers. Label the group something like “Skin Retouching” or something similar, depending on your project.
Let’s start with making adjustments to the low-frequency layer. Hide the top layer or high-frequency layer; you only see the low-frequency layer you’re going to edit. On this lower layer, we’re going to add a blur.
Navigate your way through the Filters menu to Filter>Blur>Gaussian Blur. When you apply this, you’ll get a pop-up box to adjust the blur you’ll be adding. Change the radius to make it appear that the skin is blurred most. This will typically be somewhere between 5-10px but gauge it appropriately, so there’s not too much blur.
The type of blur is not so crucial to the process. You’re free to use a Surface Blur if you want, but a Gaussian Blur should suffice just fine. All that matters is that the image has the skin blurred and will be ready for blending.
Now we’ll move onto the high-frequency layer. Display the high-frequency one above, and make sure you have this layer selected before continuing. For this layer, we’re going to use the Apply Image tool to blend the photos.
To apply this, you’ll need to navigate to Image>Apply Image. Much like with the Gaussian Blur, this will open up a dialog box. Unlike the Gaussian Blur, the settings here are a little more specific, and you’ll need to tweak some settings.
In the Apply Image dialog, set the Layer to our low-frequency layer, Blending to Subtract, Scale to 2, and Offset to 128. When this is applied, you’ll notice that the image will turn a gray color with edges of the face outlined. Don’t worry; this is what you’ll want to see.
If you’re wondering what you just did with this, you have blended with the two layers and subtracts one of them from the other. The Scale of 2 halved the effect, so it wasn’t too harsh. The 128 Offset is about half of the maximum offset and brightens up the image.
Once you’ve applied the Apply Image adjustment to the top layer, you only have one last step. Look in the layer window for the Blending dropdown box and select Linear Light from the many blending options. This should revert your image to make it look more like what you’re expecting in terms of color.
You may not notice any difference between the photo itself and the layers you just made. This is normal, and you shouldn’t see much of any change. All that you’ve done is separated different layers for making specific adjustments. Simply put, you’ve set the table for making frequency separation edits.
How To Tweak Frequency Separation in Photoshop
Now that you’ve separated the different frequencies of your photograph, you can start making changes that affect one area more than the other. You can see the changes right away by adjusting the top layer’s opacity to see more of a blur in the image. If you don’t see any difference, you may have performed the separation wrong.
This would be a great time to rename your layers, so it’s clear what you’re using them for. The top layer will be what you’ll use for Skin Texture, while the bottom layer will be Shadows & Color. Rename them accordingly so you’ll have better organization.
It’s up to you which layer you’ll be using most, but chances are high you’ll be doing more work with Shadows & Color. This layer will receive a lot of work, considering that shadows and colors can be given cosmetic tweaks while the skin texture layer will mostly blend.
Smoothing The Skin
So, what kinds of changes can we make to these layers to start seeing some significant changes in photographs? Let’s start with a common issue that most portrait photographers try to solve: reducing the rough or porous skin seen on the face.
Naturally, the solution to such rough texture realizes in smoothing out these lesser areas. Start by selecting the Shadows & Color layer. Now we’re going to isolate the rough area with the lasso tool, drawing around the parts of the face that will have the roughest spots.
Once you’ve drawn around the area, you’re going to want to feather the shape. This will prevent a harsh falloff of the area you’re affecting, so the skin looks more natural rather than just having one soft spot that looks too obviously Photoshopped. Set the feathering in the range of 30px-50px.
If you want to see just how well the feathering is looking, you don’t have to wait until the changes are made. Simply apply what’s known as a Quick Mask by pressing the Q key. It’ll show you just what part of your portrait is being sectioned off from the rest. Turn it off once you’ve got a good look.
Now we’re going to blur up that lassoed section. Navigate through the Filters to Filter>Blur>Gaussian Blur and start tweaking the levels of blur to be added in the Blur window. A good range for making such changes to the skin would be somewhere between 10px-20px for the radius.
Once you’ve applied this filter, you should notice quite the difference in the affected area. Usually, the Gaussian Blur filter will apply to the whole image but will only affect the spot you’ve chosen. If the blur doesn’t look quite right or you want to use this filter for other spots, simply undo and repeat the same process with altercations.
You’ll notice, however, that the texture will still be present. As you may have guessed, it’s because we’re only working with the shadows and color and not the textured layer. For this reason, the blur won’t be as harsh or obvious, meaning our pour-smoothing will be subtler.
How about those wrinkles on the face? It’s another common request among many who want their shots to look their most glamorous. The more conscious individuals usually want to hide this aspect of age if you’re shooting for something with beauty as a focus.
Of course, wrinkles can’t be completely removed with this process of Photoshop frequency separation. Aside from the frequency separation simply not being that powerful, the de-aging of photographs tends to make people look artificial and fake, rarely fooling anyone even with the best touch-ups.
We can, however, try to reduce the wrinkles and make them less evident through some simple techniques. This method isn’t too tough to master. It utilizes much of the same process we just used for smoothing out rough areas.
Let’s start by repeating what we’ve just done but doing it for wrinkles. With the Shadows & Color layer selected, use the lasso tool, select the wrinkles, add some feathering, and then apply the Gaussian Blur filter, tweaking to get it looking just right. The same range should work just fine but use your eyes to make the right choice.
You should start seeing some wrinkles being smoothed out but may notice some deeper wrinkles are still present. Some of them may still come through, still makes considering the texture layer is still making most of those wrinkles visible. To get around this problem, we’ll have to lighten the shadows.
After applying the blur, select the Dodge tool from your toolbar. Look at the top toolbar for tweaking the Dodge brush and change the settings for a soft brush. In the range box, select Shadows. For the exposure percentage, set it to 10%.
Armed with this tool, start painting over the wrinkled areas. If your lassoed section is still present, you shouldn’t have to worry too much about the falloff, but it’s still good to have a soft brush for larger areas. Apply as much or little as you like for toning down the wrinkles.
An alternative you can favor or use in conjunction with Dodge is the Blur tool. Similar to the Dodge tool, you’ll want to make this a soft brush as well. Soften up the brush options the same you did with Dodge. For the Mode option, select Lighten. Set the strength at 10%.
Use the Blur tool the same you used Dodge on the affected wrinkled areas. It’s up to your artistry and preference just how much looks good. Keep the undo options handy for testing out what looks good.
Now let’s tackles something a bit more complicated with blemishes on the skin. This is perhaps the more complex of aspects to remove because they’re the most visible of facial issues. Thankfully, some Photoshop techniques can clean up most of these blemishes.
Much like with wrinkles, removing blemishes is a delicate process that varies from photo to photo. You’ll have to test out what works best and make sure you don’t apply too much fixing or risk your photo looking absurdly fake.
This is the part of the process where our texture layer will come in handy. With that layer selected, bring out the Healing Brush tool and paint over some of those blemishes. You can apply the lasso tool to section off an area if you wish, but it shouldn’t be necessary if you’re handy with the brush and its specs.
If you’re comfortable enough using the Spot Removal tool, you can use that as well, although it is a bit lacking in the amount of control you can wield. No matter which tool you use, make sure the brush is soft, with the hardness at 70% and the spacing roughly 30%. The size of the blemish will ultimately determine the size of the brush.
To use the brush, you’re going to hold down on the Option/Alt button to select an area where the skin is clear, preferably near the blemish. You can then start painting over the affected area to cover up the blemish as best as possible.
Those familiar with the Clone Stamp tool will be familiar with this process as it’s quite similar. The difference in using the Healing or Spot Removal brush on this layer is that it won’t affect the overall color or shadows of the portrait.
Lightening Up The Shadows
If the shadows around the eyes or hair are looking too dark and you want to lighten them up, the process is similar to removing blemishes. The big difference here is that we’ll be using the Dodge tool on the Shadows & Color layer instead of the Texture layer.
With the Shadows & Color layer selected, select the Dodge tool and prepare a softening brush. Change the Range mode to Shadows and set the exposure at roughly 10%. Now you’re ready to start painting around those shadows that need to be less highlighted.
While painting over the areas, take care as to how much lighting needs to be applied. As with all brushes, if it doesn’t look as pronounced with the first stroke, go over it again to turn up the results. Use caution, however, as too much of the Dodge tool can create an artificial look to the dark areas.
Similar to the blemish technique, you can also use the Blur tool if you wish. If you plan to use the Blur tool, change the Mode to Lighten and the strength to 10%. However, apply your Blur or Dodge tool lightly as you want these changes to be rather subtle.
Frequency separation is not as tricky as it may sound. It’s simply a matter of splitting layers, blending those layers, and then applying changes with controlled settings. It’s a relatively straightforward process that should become a standard.
The key is to keep the photograph looking as believable as possible, so it doesn’t just look like you painted all over the problem spots. That’s essentially what you’re doing, but the aid of blurs and blendings can create a much more desired and professional look.
Frequently Asked Questions
Frequency separation essentially refers to the detailed information that is present within a photograph. There’s the high frequency which refers to surface-level features of skin imperfections and hair textures. There’s the low frequency which refers to the extra elements of shadows, color, and tone. If both are separated, they are easier to manage in editing your photo by affecting one frequency but not the other.
Splitting the frequencies in Photoshop is simply a matter of delicate duplicate layers that blend over your original image. You create duplicates of the original photo to layer on top for the high and low frequency. One will be used for skin texture, while the other will be for shadows and colors. These layers, placed on top of the original, can be blended and altered for specific changes to a photograph.
When applying frequency separation to your photographs, the key things to keep in mind are layers, filters, and brushes. For the layers, you’ll need a separate group of photograph duplicates to act as the high and low frequencies, as well as some choices in blending options. For the filters, you’ll need to blur parts of the image to smooth out rough areas. You’ll need to use the Dodge and Blur tools for fixing wrinkles and blemishes for the brushes.
The best way to think of frequency separation is to look at it is as toppings. We don’t want to change the photograph as we want to layer different parts we want to change. We want to lighten up shadows but not overexpose the skin. We want to fix wrinkles, but we don’t want the skin’s overall structure to be entirely altered. Separating these aspects into different layers that blend and blur together create a far better effect than simply applying cosmetic changes over one layer, making inconsistencies easier to come by when making broad changes.
Mark McPherson has been working as a video editor and content writer for over ten years. His background started in animation and video editing before shifting into the realm of web development. He also branched out into content writing for various online publications. Mark is an expert in video editing, content writing, and 2D/3D animation.