Why does the camera see the scene differently than the human eye?

The camera does not interpret the scene the same way the human eye does. The sensor records a specific range of brightness and color, and the final image additionally depends on exposure, optics, camera settings, signal processing, and later display.

A person can adapt their vision to changes very quickly. We can look at a bright LED wall, and a moment later at the face of the host and still relatively well read both elements. The camera, on the other hand, must work with specific exposure parameters.

If the screen behind the host is very bright, and the face is poorly lit, the video producer faces a problem: exposure set for the face may cause loss of detail on the screen, while exposure set for the LED wall will leave the person too dark.

Contrast works similarly. A very strong backlight can look impressive to the audience, but with insufficient light on the face, the camera may show the performer almost as a silhouette.

That is why lighting for cameras is not evaluated solely by looking at the scene from the audience. 

How to ensure proper exposure for the face?

The face should receive a sufficient and as stable as possible amount of light relative to the other elements in the frame. However, there is no one universal brightness value suitable for every event.

The level depends, among other things, on:

• camera settings,
• brightness of the LED wall,
• scenography,
• used optics,
• nature of the event,
• planned contrast,
• number of cameras and angles.

The relationship is the most important.

If there is a very bright screen behind the speaker, the light on the face must be appropriately balanced with it. If the background is dark, the same front level may look too aggressive.

The problem also arises when the host moves around the stage.

One well-lit position at the podium is not enough, if during the presentation the person moves a few meters to the side. It is necessary to plan the light zones so that the exposure does not change abruptly with every step.

In a panel discussion, the situation is even more demanding. Several people located in different parts of the frame should have similar exposure conditions, even if they are sitting on a wide stage. 

The direction of light and modeling the face

A good camera image depends not only on the amount of light, but also on the direction from which it falls. Light directed exactly from the axis of the camera can show the face well, but often flattens its features and the entire space.

On the other hand, a source set very high can create deep shadows in the eye sockets and under the nose.

Therefore, light on people is often set at an angle relative to the camera axis and combined with other directions.

The effect also depends on the height and width of the stage.

If the sources are very far to the sides, one side of the face may be significantly brighter than the other. In a controlled image, this may be desirable, but in a multi-camera conference, it can hinder maintaining a similar appearance from different positions.

Lighting for cameras should therefore be evaluated in the context of actual frames, not just one observation point at FOH.

Key, fill, and backlight – tools, not a rigid rule

Key, fill, and backlight are a useful way of thinking about the functions of light, but they are not a mandatory scheme that must be replicated exactly at every event.

Key light is the main direction for modeling the face or figure.

Fill light reduces the contrast created by the key and allows controlling the depth of shadows.

Backlight works from behind, helping to separate the person from the background and highlight the silhouette.

In a classic interview setup can be relatively easy. On stage, the situation is more complex: the host moves around, the panel has several people, and the cameras are in different locations.

One source can therefore be the key for the central camera, but from the perspective of the side camera, it behaves more like a side light.

Therefore, one should not only ask:

“Where is the key?”

It’s better to check:

“How does the face look in each key shot?”

That’s why lighting for multi-camera production requires collaboration between the lighting and video departments.

How to control contrast?

Contrast must be controlled as a relationship between the brightest and darkest elements relevant to the image.

This does not mean that the scene should be evenly bright.

On the contrary – lack of contrast can make the image flat and devoid of depth. The problem arises when such contrast cannot be utilized by the camera as intended.

Most often, one needs to analyze the relationship between:

• the face,
• the clothing,
• the background,
• LED wall,
• backlight,
• elements of the set.

Example: the host in a dark suit stands in front of a very bright screen. The face itself may be properly lit, but the silhouette begins to blend with the dark parts of the scenery.

Then a subtle backlight or side light may be the solution, rather than increasing the front.

On the other hand, a very strong backlight can create an overexposed outline around the hair and shoulders. From the audience, it may look good, but in close-up, the camera begins to dominate over the face.

Contrast control is best performed on monitors and measurement tools of the video path, rather than solely "by eye".

CCT and camera white balance

CCT, or correlated color temperature, describes the correlated color temperature of light. The camera must receive an appropriately set white balance for neutral surfaces to be reproduced as intended.

The problem arises when different sources with clearly varied characteristics mix on one person.

For example, the speaker may receive neutral key light from the front, cooler light from the LED wall, and another source from the venue's installation.

The human eye partially adapts to such a situation. The camera will show the differences more consistently.

This does not mean that all sources on stage must have the same color temperature.

A deliberate combination of a warmer front and cooler backlight can look very good. However, it is important that it is designed consciously, rather than resulting from an accidental mixing of sources.

White balance should therefore be established in consultation with the lighting, rather than only after the entire project is completed.

CRI, TLCI, and TM-30 – what do they say, and what do they not say?

CRI, TLCI, and TM-30 describe different aspects of color reproduction, but a single high score does not provide a complete answer to the question of how a specific source will look in a specific camera.

CRI is a common indicator for assessing color reproduction by a light source. TLCI was developed for television applications, and TM-30 provides a much broader description of the source's behavior, including fidelity and gamut.

However, the most important thing is to understand the limitations of the indicators.

ARRI points out that CRI and the fidelity index Rf from TM-30 are based on human perception and do not directly translate to how the camera sensor records color. The manufacturer even develops source modes customized to the characteristics of its ALEXA sensors, which clearly illustrates the difference between "good color for the eye" and the behavior of the source in the camera path.

Therefore a high CRI is valuable information, but it is not a guarantee of an ideal image..

For important productions, the best verification remains testing a specific fixture with the cameras used during production.

Why can LED light flicker in the camera?

LED may appear stable to the human eye, yet in the camera it can cause flicker, flashing, pulsing, or horizontal bands. The reason is the way the source is controlled and its relationship with the image sampling time by the camera.

The problem often only reveals itself after changing the camera settings.

PWM

PWM, or pulse-width modulation, is a method of regulating brightness by rapidly turning the source on and off.

If the frequency of this process does not work well with camera parameters, the sensor may register subsequent frames or fragments of a frame at different phases of the source's operation.

To the eye, the light may still appear completely continuous.

That is why fixtures designed for camera applications offer appropriate modes or the ability to adjust the behavior of the source.

Frame rate

Frame rate defines the number of frames recorded per second.

Changing the frame rate may reveal flicker that was not visible at the previous setting.

This is particularly important for slow motion and other applications requiring a high frame count.

Therefore, the test "everything is fine on a standard camera" does not guarantee that the same fixture will behave identically when switching production to different parameters.

Shutter angle / shutter speed

Shutter angle or shutter speed defines the exposure time of a single frame.

Changing the exposure time affects how the camera attempts to capture light pulsing at a specific frequency.

Thus, there may be a situation where at one setting the image is clear, and after changing the shutter, bands or flickering appear.

Therefore, the flicker issue should not be assessed solely based on the lamp's specifications.

One must analyze the lamp and camera settings as one system..

High-speed/flicker-free modes

Some fixtures have modes developed specifically for applications requiring particular resistance to flicker.

ARRI describes in the SkyPanel documentation a special High Speed Mode designed for high frame rate production. The manufacturer also indicates that at high frame rates or extreme shutter settings, one can use High Speed Mode or appropriately adjust the control frequency of the source.

However, this does not mean that the label "flicker-free" exempts from tests.

Different modes may have their own limitations regarding for example dimming or fixture functions. Therefore, the setting must be tailored to the specific recording method.

How to reconcile the LED wall with face exposure?

The LED wall and face should be within such a range of brightness that the camera can simultaneously maintain a clear image of the person and necessary details of the content.

This is one of the most common event problems.

Behind the presenter is a large screen with a bright presentation. For the audience, everything looks good, but in the camera, the screen begins to dominate over the face.

The first reaction is often to increase the light on the presenter.

Sometimes this is necessary, but there is also a second tool: reducing the brightness of the LED wall.

This can improve the ratio without unnecessarily increasing the light level on the individuals.

One must also be careful with the content.

The same brightness of the LED processor does not mean the same perception of each graphic. A large white surface and a dark animation behave visually completely differently.

Therefore, during the rehearsal, it is worth seeing actual presentations, videos, and boards, not just the test screen.

With the integration of lighting, multimedia, and sound lighting and video should establish this relationship before the event begins.

Why does "adding more light from the front" not always solve the problem?

More front light can improve exposure, but at the same time ruin facial modeling, reduce contrast, and flatten the entire scene.

If the problem is a very bright LED wall, adding more levels of front light leads to a kind of "brightness race".

The screen is bright, so we increase the front. After increasing the front, the scenery appears darker, so we increase additional sources. As a result, the entire scene operates at a much higher level, even though the original problem could have been solved by reducing the brightness of one element.

Similarly with strong backlighting.

If the contour is too bright compared to the face, you can increase the key, but you can also decrease the backlight. The second solution can maintain much more natural proportions.

Sometimes the problem arises not from the level, but from the direction. There is plenty of light, but it falls at an angle that creates unfavorable shadows.

The design for cameras is therefore about balancing the system, not just adding more lumens.

What does a camera test look like?

The camera test should be conducted with real cameras, lenses, and settings as close as possible to those used during the event.

It is best to check not an empty scene, but a person standing in real positions.

During the test, it is worth assessing:

• the exposure of the face,
• the differences between areas of the scene,
• the balance with the LED wall,
• the colors of skin and wardrobe,
• contrast,
• backlight,
• behavior of the fixture during dimming,
• flicker,
• the most important camera frames.

In multi-camera production, each main position needs to be reviewed.

The look may appear great from the central camera, but the camera set at an angle shows the same person from the shadow side.

Video tools such as waveform are helpful, allowing for a more objective assessment of signal levels than just the monitor preview.

The test should also include changes to the LED content.

If both dark films and bright presentations occur during the event, it is necessary to check both extreme cases.

Concert, gala, and conference – three different cases

Concert, gala, and conference require a different compromise between the image for the audience and for the camera.

Concert can utilize significantly greater contrast. A dark face in one part of the piece can be an intentional artistic move, similar to silhouette, intense backlight, or a momentary blackout. The camera should then reflect the intended drama, rather than forcibly equalizing the entire image.

Gala connects two worlds. The host and the award recipients should be clearly visible, but openings, entrances, performances, and presentations can benefit from more spectacular lighting. Moments intended for photography and close-ups are particularly important.

Conference usually requires the greatest stability. Speakers moving around the stage should remain legible, and the display should not change abruptly when transitioning between positions.

Dlatego lighting for conferences and corporate events should take into account the camera plan during the design of the front zones.

How do the lighting designer and video operator collaborate?

The lighting designer is responsible for stage lighting, while the video operator is responsible for the camera image, but both work with the same material – the stage in front of the lens.

Therefore, decisions made by one party directly affect the other.

The lighting should be aware of, among other things:

• the positions of the cameras,
• the most important shots,
• settings regarding frame rate,
• requirements for slow motion,
• the way LED wall is used,
• the nature of the production.

The video, on the other hand, should convey information about issues with exposure, color, flicker, or contrast.

It's not about subordinating the entire lighting design to the cameras.

At a concert, the lighting can still be very contrasting and artistic. The video operator simply needs to know what effect is intended.

The worst situation arises when both departments work independently and only meet a few minutes before the broadcast.

A joint rehearsal allows for establishing the proportions between key light, the background, LED wall, and backlight, as well as checking problematic LED sources.

For a broadcasted event the joint project of lighting + video/streaming allows treating the stage and camera image as two aspects of the same production, instead of trying to fix the exposure only at the video console.

Summary

A scene that looks good for cameras does not have to be flat or evenly lit. It primarily needs controlled exposure, conscious directions, appropriate contrast, and the right relationship between people, the set, and the LED wall.

Key, fill, and backlight are useful tools, but they should be matched to the actual frames. CCT needs to be coordinated with the white balance, and indicators such as CRI, TLCI, or TM-30 should be treated as information about the source, not a substitute for a camera test. ARRI points out that metrics based on human perception do not directly translate to the response of a specific sensor.

The most important rule is simple: light for cameras must be evaluated in the camera. A joint lighting and video test allows for correcting the exposure, LED wall, colors, and flicker before the broadcast begins.

Frequently Asked Questions