How to Use a Color Chart in DaVinci Resolve (and Mistakes You Should Avoid)

How to Use a Color Chart in DaVinci Resolve (and Mistakes You Should Avoid)

What exactly is the purpose of a color chart and how can we use it to get better grades? It’s a simple premise, once you understand how color charts work and how the Color Match feature works inside DaVinci Resolve. But there are a couple of plot twists you’ll want to watch out for. In this article, I’ll cover how I take advantage of this feature.

Understanding the color chart

Before we dive into the Color Match feature itself, I want to explain what color charts are used for. Whether we’re using the Color Match feature in Resolve or for any other purpose, why do we shoot color charts in the first place? This is something that took me a little while to understand as a colorist.

The image looks good, but we don’t know what the original colors in the scene should be.
This image looks good, but we don’t know what the original colors in the scene were.

Take a look at the image above. We don’t actually know what the original scene colors in this image should be. We can see how the colors are being reproduced and make some reasonable guesses, but we don’t numerically know what all those colors should be. We don’t know exactly what they were in the scene on that day.

But, because we have a color chart in our frame, Resolve knows exactly what the numerical values of the colors and grayscale steps in this color chart should be.

DaVinci Resolve uses the Color Matching tool to determine the precise shade of the color chart.
DaVinci Resolve uses the Color Match tool to determine the precise shade of the color chart.

Because Resolve stores these reference color values internally, it can use that information to compare what these colors should be versus what was captured and encoded. Resolve can then calculate a match to lessen the distance between what these color patches should be and what they are.

By doing this, we are theoretically going to get a more balanced image with more accurate color reproduction. This can be a useful foundation when we begin our color grade.

Using the Color Match tool

Here’s how you can use Resolve’s Color Match feature to take advantage of the utility of a color chart. First, in the lower left-hand corner of Resolve’s Color workspace, select the Color Match palette.

DaVinci Color Match palette menu select
Open the Color Match palette…
Choosing Color Chart in the drop-down menu will bring up the palette tool.
…then select the Color Chart option from the drop-down menu to bring up the overlay.

Next, select the Color Chart option mode from the dropdown menu in the lower left-hand corner of the viewer to bring up the color chart overlay. Fit the overlay so that each inner patch is filled by its respective square within the color chart.

This overlay tells Resolve which shades on the color chart to match to their corresponding numerical value.
Adjust the overlay so that the squares sit on top of each of the color patches.

Now we have to tell Resolve what kind of color chart we’re looking at. But how do you know if this is a ColorChecker Classic? As you can see, there are a few different options that we can select from the drop-down menu.

There are many color charts to choose from.
X-Rite’s ColorChecker Classic is a common color chart for both photo and video color workflows

In this example, we know that our chart is a ColorChecker Classic because, if we zoom in, we can see the name of the chart in the upper left-hand corner.

The name in the upper left-hand corner tells you which chart you have.
We can see that our chart is a ColorChecker Classic thanks to the label in the corner.

ColorChecker Classic is one of the most common charts you’re going to encounter when you’re using the Color Match feature. In general, it’s one of the more popular charts used on sets to capture known references for seeing colors.

Notice that there are two ColorChecker options: the ColorChecker Classic and the ColorChecker Classic – Legacy. If you don’t know which chart you have, I’d encourage you to go with the ColorChecker Classic. If you go out and buy a ColorChecker chart today, that’s the one you’ll get. The difference between the two is quite minor anyway. Either way, you’re still going to get a better foundation than if you didn’t use a color chart at all.

Choosing your configuration settings

Before we tell Resolve to do its automatic bit, we need to specify a few things. We need to clarify what we’re trying to do.

Remember, we’re working inside of a color-managed pipeline right now. So in this example, our footage is going from the camera color space (ARRI Log C4), moving into the working color space (DaVinci Wide Gamut Intermediate)—that’s where we’re going to do all our color grading—and finally out to our display color space (Rec. 709, Gamma 2.4).

Before hitting Match, DaVinci Resolve has to know what color space it is working in.
Before hitting Match, Resolve has to know what color space it is working in.

How does what we’re about to do fit into that pipeline? We’re looking for a transform that aligns the colors of the ColorChecker to what they should be. This transform should minimize the error between the colors as they should have been captured versus how they were captured.

We want to do all of that within our working color space. We don’t need to change color spaces. Everything should happen within DaVinci Wide Gamut Intermediate. That’s going to inform the way that we set up these next couple of drop-down menus.

Setting up your configurations ensures you’re working in the correct color space.
Setting up your configurations ensures you’re working in the correct color space.

In the drop-down menus, we’ll set our Source Gamma and Target Gamma to DaVinci Intermediate, and our Target Color Space to DaVinci Wide Gamut. Because we’ve set things up in this way, everything is going to happen inside of DaVinci Wide Gamut Intermediate.

Here, I’m working on my Balance Node in my node tree template. You can choose to work in a separate node if you want to, but this is a perfectly fine place for us to run this Color Match.

The Balance Node is a good place to run a color match.
The Balance Node is a good place to run a color match.

Now let’s take a look at what happens after we hit Match.

Our image before the color match…
Our image before the color match…
…and after, with the newly-calibrated colors.
…and after, with the newly-calibrated colors.

It’s nothing earth-shattering because this is a well-shot image. It was shot under properly calibrated lighting conditions.

Now’s a good time to point out one important detail: the color chart in this image was shot under the same lighting conditions as the scene. If you have a chart that was lit using a different light than the scene received, then the color match will not work the way you want it to. You’ll get a perfectly calibrated set of colors on your chart, but that will not translate to the colors of your actual scene—the whole point of the ColorChecker.

We don’t care about seeing the color patches themselves reproduced accurately. We care about them providing a reference for getting a more accurate reproduction of the scene itself. You want to make sure your ColorChecker was lit by the same light that was lighting the scene.

“We don’t care about seeing the color patches themselves reproduced accurately.”

This is a mistake that often happens on set. Sometimes, a color chart is lit up by a camera assistant’s flashlight or some other secondary light source. But if you want to use the chart to get better reproduction of all your scene colors, then it needs to be lit by the same source that is illuminating the scene itself. Otherwise, you’re just getting a reproduction of the ColorChecker chart itself.

Applying a color match to the image

Two things have happened as a result of hitting that Match button. First, we’ve remapped those hues. We can see that things in the scopes have moved around.

Comparing the histogram before…
Comparing the histogram before…
...and after will show how Color Match has changed the image.
…and after will show how Color Match has changed the image.

Resolve has aligned the hue and saturation of the color patches to what it knows they should have been if they had been captured by a perfectly accurate camera under ideal lighting conditions. Resolve is fitting what was captured into what should have been captured by a theoretical, perfect system.

But something else has happened, hasn’t it? The image has dropped in exposure, just a hair. The color match has made it just a little bit darker.

A side-by-side comparison shows the dip in exposure after the Color Match.
A side-by-side comparison shows the dip in exposure after the Color Match.

Our image is getting darker because the ColorChecker believes that is what we need to do to match our gray patches. In particular, the gray square is being dropped down to something closer to 18% gray, in linear terms.

A side-by-side comparison of the color chart. Note the gray squares at the bottom are a touch darker.
A side-by-side comparison of the color chart. Note the gray squares at the bottom are a touch darker.

Things to look out for while color matching

Now that we know what a color chart is and how to make use of it inside of Resolve, let’s take a look at a few things to watch out for.

Another example of a color chart, in a scene lit with an interior night atmosphere.
Another example of a color chart, in a scene lit with an interior night atmosphere.

In this new example, we need to pay particular attention because there’s an occlusion that we have to work around. As you can see, a silver C-Stand knuckle is cutting into the upper left corner of our color chart.

Objects cutting into the color chart, like this C-stand knuckle, can throw off the color match.
Objects cutting into the color chart, like this C-stand knuckle, can throw off the Color Match.

We have to make sure that we’re not getting any of that knuckle inside our color sample region because that will skew the analysis. We don’t want any of those pixels to be included in the assessment of where the color is encoded onto the sensor.

After we’ve made sure that the little square is nothing but the color patch itself, we have to choose our Source Gamma, Target Gamma, and Target Color Space, just like we did before.

Make sure your configurations are set correctly before each color match.
Make sure your configurations are set correctly before each Color Match.

By the way, it’s perfectly fine to leave the color temperature at 6500 Kelvin, because that is the native white point of the DaVinci intermediate space that we are working in. Now, let’s hit Match.

The color difference is apparent immediately.
Compare the two color charts and you can see a significant difference.

As you can see from the differences between the color chart before and after, Resolve arrived at a different solution here than in the first shot, but we’re doing a good job of getting things back to neutral.

Making the color chart work for you

And that brings up an interesting point that I want to emphasize. Sometimes if we’re looking at pure aesthetics, we might prefer the original image over the adjusted image.

Compare the result of our color match adjustment…
Compare the result of our Color Match adjustment…
…with the original image.
…with the original image.

If I toggle between these two images, I might personally prefer the original image over the adjusted image. Maybe you do too.

So, when should we use a color chart and do we have to use the chart if it’s in the frame? The answer is no, we absolutely do not have to use it.

Here’s the prescription I use when it comes to color charts: If you have a chart and it is shot under lighting conditions that match the scene, then go ahead and run the color match just to see what it gives back to you. You can choose to either take a little bit of it, you can toss it out completely, or you can say, “This gives me a better, more balanced starting point for my image.”

“You don’t have to use all or even any of your color match.”

You don’t have to use all or even any of your color match once you pull it, but now you have the advantage of seeing the image balanced to the colors in the color chart. You can see what the image looks like if the color chart was captured under perfect lighting conditions by a perfectly accurate camera.

With an image like this, splitting the difference is also an option. We can use our Key palette to set the Key Output Gain to 0.5.

Lowering the Key Output Gain can decrease dramatic balance changes…
Lowering the Key Output Gain can decrease dramatic balance changes…
…and give a more pleasing image.
…and give a more pleasing image.

Now we’re only getting half of that color rebalance. You can have all of that color match, none of it, or just a percentage of it.

Always remember the intention

But what are the rules? How do you know what to do when? Ultimately, it just comes down to your eye. It also comes down to the creative intent of your collaborators. Maybe the warm look that we started with is the exact intent of the filmmakers.

This intent is something else that you want to be sure of. As a colorist, you want to discuss artistic intent with the people that you’re working with. If your collaborators see that you’ve balanced everything to be neutral, but they intended the image to be warm, then they are going to be surprised. They’re going to say, “What did you do?”

I want to be clear: the ColorChecker is there to support you, but you don’t owe anything to the ColorChecker. It’s there to work for you. If it is working, then you should use it. But if it’s not working for you, feel free to throw it away.

Now you know how color charts work in general and how the Color Match feature works inside of DaVinci Resolve. You can choose when and how you might want to use it to get better grades for your images.

Thank you to Cullen Kelly for contributing this article.

Cullen Kelly is a Los Angeles-based senior colorist with credits spanning film, television, and commercials, for clients and outlets including Netflix, HBO, Hulu, Microsoft, McDonald’s, and Sephora. With a background in image science as well as the arts, he’s passionate about the intersection of the creative and technical, and its role in great visual storytelling. In addition to his grading work, Cullen is an educator and proven thought leader, with platforms including his podcast The Color Code as well as his YouTube channel.

Interested in contributing?

This blog relies on people like you to step in and add your voice. Send us an email: blog at frame.io if you have an idea for a post or want to write one yourself.