Many tutorials for the Frequency Separation technique involve a few steps that I feel are unnecessary. I would like to apologize in advance if I do not use the proper terminology...
They usually involve starting by making two duplicate layers (high and low frequencies), turning off the high while finding the best blur radius, then using Apply Image to subtract on image from the other to regain the texture.
A faster method I have found is to make one duplicate layer, using your choice of blur (Gaussian or surface blur) inverting the layer, setting the opacity to 50%, then stamping the image to a new layer set to Linear Light blend mode. Go back to the layer you inverted and set the opacity back to 100% and invert it back.
The idea behind this is if you were to take the duplicated layer (without any blurring applied to it) and inverting it, the colors become the inverse; at 50% opacity, the inverted colors cancel everything out. If you apply a Blur to the layer, you essentially created a high pass layer because it's the inverted layer is the inverse of the blur.
To help find the blemishes/imperfections of skin: create a Channel Mixer adjustment layer and set it to monochrome. Move the red slider to -100, the green channel can stay at 0 or 100 (play with it), and set the blue slider to +100 or +200. You can also set the red slider to +100, the green to +100, and the blue to -100.
Now you can create a mask by subtracting one layer from the other to target the areas you will need to retouch.
Hopefully this info makes sense...
Sent from my iPhone using Tapatalk
They usually involve starting by making two duplicate layers (high and low frequencies), turning off the high while finding the best blur radius, then using Apply Image to subtract on image from the other to regain the texture.
A faster method I have found is to make one duplicate layer, using your choice of blur (Gaussian or surface blur) inverting the layer, setting the opacity to 50%, then stamping the image to a new layer set to Linear Light blend mode. Go back to the layer you inverted and set the opacity back to 100% and invert it back.
The idea behind this is if you were to take the duplicated layer (without any blurring applied to it) and inverting it, the colors become the inverse; at 50% opacity, the inverted colors cancel everything out. If you apply a Blur to the layer, you essentially created a high pass layer because it's the inverted layer is the inverse of the blur.
To help find the blemishes/imperfections of skin: create a Channel Mixer adjustment layer and set it to monochrome. Move the red slider to -100, the green channel can stay at 0 or 100 (play with it), and set the blue slider to +100 or +200. You can also set the red slider to +100, the green to +100, and the blue to -100.
Now you can create a mask by subtracting one layer from the other to target the areas you will need to retouch.
Hopefully this info makes sense...
Sent from my iPhone using Tapatalk
Last edited: