Tutorials Motion

Matching Shots From Different Cameras

By Infinite Creation · Intermediate · ~18 min

A 1960s mid-century comic book style editorial illustration showing a motion designer working at a large monitor, with a camera on a tripod, a colour grading panel with trackballs, a vectorscope and waveform monitor visible in the frame.

Overview

Multicamera footage that does not match is one of the most visible signs of an amateur production, and audiences detect it instantly even when they cannot say why. Different sensors, different profiles, and different white balance settings mean two cameras pointed at the same scene rarely produce the same image. Matching them is a systematic process with a fixed order, not an exercise in eyeballing. This guide covers that order, and what to do on set to make the job small.

What You Need

Footage from two or more cameras of the same scene
Scopes: waveform, parade, and vectorscope
A grading tool that lets you copy grades between clips
A colour chart or grey card shot on each camera, if you have one
A reference clip chosen deliberately as the target
A screen at fixed brightness in consistent room lighting

Steps

Shoot a chart on every camera before you roll

A few seconds of a colour chart or grey card, lit by the same light, on every camera, at the head of each setup. This costs almost nothing on set and turns matching from a judgement exercise into a measurement one. It is the single highest-value thing you can do for the grade, and it is the step most often skipped.

Normalise everything to a common starting point

If cameras shot in different profiles (one log, one standard) convert them to a shared colour space before comparing anything. Trying to match a log clip against a Rec.709 clip directly is fighting two problems at once. Colour management handles this properly. A conversion LUT is the manual equivalent.

Choose one camera as the reference and grade it first

Pick the camera with the best exposure and the look closest to your intent, grade it properly, and treat it as the target. Matching everything to a fixed reference is far faster than adjusting several clips toward each other simultaneously, which tends to oscillate without converging.

Match in a fixed order: black, white, mid, colour, saturation

Set the black point first, then the white point, then overall midtone brightness, then colour balance, then saturation. Order matters because brightness changes alter apparent saturation, doing colour before exposure means redoing the colour. Use the waveform for the first three and the parade for the fourth.

Use the parade for colour, not your eyes

The RGB parade shows each channel separately. When two clips match, their parades have the same shape at top and bottom. Aligning the channels at the shadow and highlight ends fixes colour casts far more reliably than adjusting a temperature slider while looking at the picture, especially after an hour in a dark room.

Check with a split screen and at speed

Compare the two shots side by side, then watch the actual cut at normal speed. Matching that looks perfect on stills can still read wrong across a cut, and matching that looks marginally off on stills frequently plays fine. The cut is what the audience sees, so it is what you judge on.

Pro Tips

Shoot a grey card on every camera at every setup. Nothing else saves as much grading time.
Lock white balance to a number on every camera. Auto white balance guarantees cameras will not match.
Match faces before backgrounds. Audiences forgive a background mismatch and never forgive a skin tone one.
If one camera is badly overexposed, match the others toward it rather than trying to recover clipped highlights.
Save your matched grades as stills or presets so later scenes with the same cameras start from a known point.

Tested: does the order of a match matter, and why exposure comes first

We tested step 4 and its Knowledge Base explanation on 2026-09-29 with a model of two cameras written in numpy. It is a model, not footage. A chart of 24 patches, six greys and 18 colours, was rendered by a reference camera and by a second camera with a white balance error, 0.4 of a stop too much exposure, less saturation, a mild contrast curve and a lifted black level. We then matched the second camera to the first with five plain steps in the page's order, each one number read from the chart the way a scope is read, and measured the colour difference left across all 24 patches, in CIELAB delta E units, where about 1 is roughly the smallest difference a viewer can see side by side.

After this step, in the page's order	Colour difference left (delta E)	Change from the step before
Before any step	9.0	start
Black point	8.5	-0.5
White point	5.7	-2.7
Midtones	5.7	0.0
Colour balance	2.4	-3.3
Saturation	1.4	-1.0

The five steps took the difference from 9.0 to 1.4. They did not take it to zero, and could not: the second camera's contrast curve bends the tones, and one gamma number cannot undo a curve. That leftover is why step 6 asks you to check the cut at speed and not only a still.

Step 4's premise holds, that the adjustments interact. We ran the same five steps in all 120 possible orders. The difference left ranged from 1.0 for the best order to 5.4 for the worst, with a mean of 2.0. Running the page's order backwards left 2.2.

The test does not show that the page's order is the best one. It left 1.4, which ranked 50 of 120, better than the mean and not the best. Orders that did colour before all three tone steps averaged 1.3, against 2.1 for orders that did it after them, the opposite of "doing colour before exposure means redoing the colour". That comes from how each step is written here, and a colourist's tools estimate differently, so read it as evidence that order changes the result and not as a ranking of orders.

The Knowledge Base's saturation point holds, with a catch for the scope. We rendered the same chart one stop darker, as shot and one stop brighter. The colours' chroma relative to their lightness fell 7 percent for the extra stop of exposure, so they looked less saturated, as the page says. The vectorscope moved the other way: the average distance of the colour patches from the centre rose 37 percent, from 0.089 to 0.121. The scope's reading of saturation goes up as the picture looks less saturated, which is a good reason to settle exposure before you trust either.

Validated in a model. Measured 2026-09-29 with numpy. The two cameras are simulated, and each matching step is our own simple estimate from the chart, so the ranking of orders depends on how the steps are written and would differ in a grading tool. The chart is synthetic, the colour difference is CIELAB delta E 1976 averaged over 24 patches, and the vectorscope reading is computed in BT.709. We did not test real cameras, log profiles or any colour management.

Knowledge Base

What You'll Learn

Matching is the largest part of professional grading work, and it is a measurement task rather than a taste one. Below: why cameras differ in the first place, and why the matching order is fixed.

Why Two Cameras Never Match Out of the Box

Even identical camera models shooting the same scene produce different images, and understanding the causes tells you which are fixable on set.

Sensor and colour science differ. Manufacturers make different choices about how sensor data becomes colour, particularly in skin tones and saturated reds. This is why footage from two brands has a recognisable character difference even at identical settings.

Profiles differ. A camera shooting a standard profile has already applied contrast and saturation. One shooting log has not. These are not comparable until converted to a common space.

White balance differs. Even set to the same nominal value, cameras interpret it differently, and if either is on auto, they will diverge continuously.

Exposure differs. Different lenses, different metering, different positions relative to the light.

Of these, white balance and exposure are entirely controllable on set and account for most of the mismatch. The colour science difference is the residue that grading exists to handle.

Why the Matching Order Is Fixed

Matching feels circular when done in an arbitrary order, and the reason is that the adjustments interact.

Brightness affects apparent saturation. Raise the exposure of a clip and its colours appear less saturated. Lower it and they appear more so. Any colour work done before exposure is settled will need redoing.

Black and white points define the range everything else sits in. Setting them first fixes the endpoints, so subsequent midtone and colour adjustments happen within a stable frame.

Colour balance is easiest to judge once brightness matches. Two clips at different brightness look different in ways that mask and mimic colour casts.

So the order (black point, white point, midtones, colour balance, saturation) moves from the adjustments that constrain others to the ones that are constrained. Following it means each decision is made once. Ignoring it is why matching sometimes feels like it never converges.

Where This Fits

Matching cameras is the correction step in Color Grading in DaVinci Resolve, which then builds a look on top that holds across every shot.

FAQ

Q: What is the fastest way to match two cameras?
A: Shoot a grey card or colour chart on both at the head of the setup, then match measured against the scopes rather than by eye. Without a chart, pick a reference camera, grade it, and match the other in the fixed order, black point, white point, midtones, colour balance, saturation.

Q: Should I match by eye or with scopes?
A: Scopes. Your monitor may be uncalibrated, the room light shifts your perception, and your eyes adapt continuously as you work, so visual judgement drifts across a session. The waveform and RGB parade do not adapt, and two clips whose parades align will cut together.

Q: Can I match a log clip to a standard-profile clip directly?
A: Not usefully. Convert both to a common colour space first, either through proper colour management or a conversion LUT, then match. Attempting it directly means fighting the profile difference and the camera difference simultaneously, which is why it feels impossible.

Q: What should I match first if the shots are very different?
A: Faces. Audiences tolerate background mismatches and notice skin tone differences immediately, so prioritise getting people consistent across the cut. If one camera is badly overexposed with clipped highlights, match the others toward it rather than trying to recover information that no longer exists.