How to use High Pass and Luminosity Passes for Blending

Learn how to use High pass and Luminosity layer blending modes for Artistic and Technical purposes with Senior CGI Generalist Tamas Gyerman

How to use High Pass and Luminosity Passes for Blending
Tamas Gyerman
Hello there, my name is Tamas Gyerman and I am a Senior CGI Generalist Artist / TD. My professional specialty is digital environment creation, both 2D and 3D. In this article you will learn how to use High pass and Luminosity layer blending modes in Photoshop. These two functions are useful for artistic and also for technical uses in your artwork, for example: paint overs, sharpening, neutralizing, material map generation, detail refinement and many other scenarios.

High Pass

Let’s look at the basics before we move forward. When you work in Photoshop or a similar software, we are working with Pixels. When you change shade, color tone, brightness, saturation, etc; these are changes made from color and image information from our Brain. However these changes are represented as actual numbers in a digital database and based on these values you can see the visual feedback/result. In general, Graphic Cards generate color based on three major components: Red, Green, Blue or RGB. Each Color is represented by a number value, for instance, the value of black is 0:0:0, and the value of white is 255:255:255.

You don't need to gamble about what you can expect, if you know how to prepare your stuff.

Here is the important part of these numbers; the mid solid gray value is 128:128:128. Why is this important for us? The reason is that when you do different layer blending modes in Photoshop; the algorithm gives you the result based on these number values of your image content. For example; 'Multiply' ignores all white values, 'Screen' does the inverse of that (ignoring the black values) and 'Overlay' does so with gray. Knowing this, you can actually plan how you prepare your image content for further usage. You don't need to gamble about what you can expect, if you know how to prepare your stuff.


Gray is the key for High Pass/Overlay, and also for Luminance blending modes. Let’s start with the High Pass and have a look at a short example. This function is available through Filter -> other -> High Pass.

The smaller the value is, the sharper the detail you get. When you increase the number, the script gives you more details but as you can see, it forces the base to keep it neutral gray. When you do 'Overlay"; as I mentioned this value (Grey) will turn invisible. Or if you don’t need any color value on your layer, just go with a gray-scale image from the beginning.

For example, High Pass is great for additional detail blending and also for refining without destroying the image with a sharpening process. You can see this in the example below:

To learn more about additional detailing, you can read more about Hard surface texturing / concept detailing at this link to see how the texture was created in the below images:


Luminosity

Let's continue with the Luminosity blending mode. In use, the Luminosity blending mode allows us to blend content on top of each other without dealing with lots of color information. The most simple way I can explain this is that the layer loses most of the color information like it has been saturated but also makes a blend with the other content. The actual layer picks up color data from the underneath content and blends the details together. Sometimes you need to force down your color information with saturation for better results.

“The Luminosity blend mode preserves the hue and chroma of the bottom layer, while adopting the luma of the top layer.”

For instance, this blending mode is great for detail painting. Imagine a castle, or a stone wall, or a tree. If you want to add details onto these surfaces, it is always a game of matching the colors. This is not the case when you use the Luminosity layer blending mode. This allows you to save a lot of time not dealing with colors, and you won’t need to destroy too much of the original content as the blending mode is done in a procedural way and it doesn’t bake the changes into the image data. Let’s see a few very simple examples:

In the above examples, the windows are from a different image. I used the underneath paint for the glass color and used the original wall colors for luminosity blending.


In the example below, a I did a quick patch to cover the metallic element under the pigeon. I used a piece of bark from another picture. The tones are different but with luminosity blending it fits well.

Another example showing how you can add details to rocks with luminosity blending.

For a more detailed and accurate explanation about blending modes, please visit the following link: Blend Modes


Bump and Displacement map values:

If you have never heard about Bump and Displacement map values, these maps allow you to generate varied surfaces from flat surfaces. With Bump you can create additional details which go into or stand out from the surface. With Displacement you can do the same, but the major difference is that Bump is only fake adjustments and from a low /straight angle you can’t see it but Displace generates actual geometry when working in 3D; like it was modeled properly.

To create high resolution, small details, like scratches, paneling, rivets, dirt, leak erosion, etc, bump is a great solution. Displace is great when we actually need the silhouette of the geometry, or we want to calculate more accurate lighting or to cast shadow. There are lots of interesting descriptions about these on the internet with much more details, for now this is just about the overall understanding. Check out these Links for more information: Bump Mapping, Normal Mapping, Displacement Mapping

Why is high pass important for us in this scenario?

Bump and Displace are all based on gray-scale information and as we previously discussed; High Pass generates content based on 50% gray values. Using this method you can create additional texture/shade information. It is very important to know that for proper calculation, you must have at least 16 bit for your texture. The reason is that 8 bit does not have enough range to create/generate smooth enough falloff. If you use it, your surface will be clipped.


Let’s circle back to resolution/bit depth. This image is a super close up, sharpened displacement render with the same map, but an 8 and a 16 bit version. As you can see, the 8 bit has a lot of surface artifacts but the 16 bit looks much smoother.

In Photoshop, when we try to create falloffs/gradients, we have a lot of steps on 8 bit, however on 16 bit everything looks much smoother. It is important to keep this in mind because when we use High Pass for Displacement map enhancement, we might end up with similar issues, because the source images are usually 8 bit Jpeg's. It does not matter if you blend on top of a 16 bit generated map, as the software doesn't always calculate enough depth fill from non existing data. In general, converting bit depth without proper precaution can cause you a lot of headache and should be avoided. Always make sure you have backups, especially with low range images such as night time.


So, in this short article we explored and discussed High Pass and Luminosity functions, and we also got a small taste about generating surfaces via textures in 3D, using Bump and Displacement maps. We also learnt a better understanding of what happens if we convert or do not use proper bit depth. If you have specific questions, please get in touch!

Thanks for reading, if you would like to see more of my work or connect with me, follow me here:

Artstation | Linkedin

~ Tamas Gyerman