How to Copy a Color Grade from a Reference Photo

Every professional colorist keeps a library of reference images. Not because they're copying other people's work — but because describing color is hard. It's much easier to say "I want this photo to feel like that photo" and let the tools handle the technical translation. Reference-based grading is how color decisions get made in almost every major film and commercial production.

What Reference-Based Grading Actually Is

Reference-based grading means extracting the color properties of a source image and applying them to a target image. The colorist picks a "look image" whose palette, contrast, and tonal character they want to replicate. The grading tool analyzes the reference — its color distribution, split-tone direction, contrast shape, saturation balance — and applies a similar treatment to the new image.

It's not a direct pixel-for-pixel copy. It's a stylistic match — the new image looks like it was shot and graded by the same person as the reference.

How to Pick a Good Reference Photo

This is where most people go wrong. When browsing for references, it's easy to pick a photo you love for its composition or subject — a beautiful landscape, a stunning portrait. But the subject isn't what you're extracting. You're extracting the colors.

• Pick references whose color palette you love, not whose subject you love
• Look for images with similar lighting conditions to your own photo (golden hour to golden hour, indoor to indoor)
• Avoid heavily composite or retouched photos — their color balance may be artificial
• Film photography is an excellent reference source — the color science is well-established
• Look at the shadows first: shadow color direction tells you the most about a grade

What Gets Extracted from a Reference Image

When an AI tool analyzes a reference photo, it looks at several properties:

• Color distribution — which hues are dominant and how saturated they are
• Tonal contrast — where the blacks land, how the highlights roll off, the shape of the midtones
• Split-tone direction — whether shadows lean cool, warm, teal, or green; whether highlights lean amber, white, or yellow
• Grain character — the texture and frequency of film grain if present
• Overall saturation level — whether the image is punchy or muted

Getting the Best Results

Reference grading works best when the reference and target images share similar base conditions. A reference shot in harsh midday sun will produce different results when applied to a soft overcast portrait — not wrong, but the translation will be imperfect. If you're matching a specific film stock or editorial look, consider combining reference grading with a text description: "apply the color palette of this reference, but add more grain and lift the blacks slightly."

How Reference Mode Works in Asterik

Asterik's reference mode lets you paste or upload any image as a visual inspiration source. The AI extracts the grade signature — color direction, contrast, saturation profile — and applies it to your uploaded photo. You can combine it with text modifiers to push the result further, and the resulting grade can be saved as a reusable filter for your entire catalog.