Headline_CompositingDepthCues

Pictorial depth cues constantly surround us, yet the human brain is so adept at making these instantaneous calculations and assumptions that we hardly realize we deciphered potentially random visual data at all. It’s only when these depth cues are misaligned or incomplete (inadvertently or not) does our perception take a step back and need to be reprocessed.

DepthCues_IntroAnimators, graphic/motion designers and filmmakers need a solid command of identifying the most common depth cues to allow their work to ‘read’ and be understood as quickly as possible– and with the least amount of viewer confusion. Most of us are unconsciously aware of these cues, but to quantize and analyze these visual phenomena will help use utilize them more effectively, and even break them for effect when desired. Let’s take a look!


Atmospheric Perspective
DepthCues_AtmoAtmospheric Perspective is the result of light being scattered by particles in the air over great distances. More distant objects take on the hues inherent in the sky at the time (blue tints for afternoon, red tints for sunsets, etc.) The contrast of any markings or details within the objects also decreases, as does the native saturation.

This shouldn’t be confused with volumetric lights that exhibit participating media in a small area – such as God Rays through a window or a flashlight beam on a foggy night. The underlying concepts are similar, but volumetric lighting does not give an indication of distance.


Overlap and Occlusion
DepthCues_OverlapThis cue is one of the most common, and is the first thing people think of when attempting to discern depth. Also known as an interpositional cue, this is the result of objects more near to the camera covering up parts of other objects further away – even when both objects are placed at the same distance.

However, this information only allows the observer to create a ranking of relative nearness – it can’t be used to determine actual size discrepancy.


Linear Perspective
DepthCues_LinearPerspThis property of parallel lines converging in the distance at infinity allows us to reconstruct the relative distance of two parts of an object or of landscape features.

This concept eluded artists through the Middle Ages until 1435 when Leon Battista Alberti provided the first theory of accurately reproducing linear perspective in his book On Painting.

A good example is how the rails on railroad tracks appear to get closer together in the distance.


Relative Height
DepthCues_RelativeHeightA monocular cue in depth perception and distance in which higher objects in the two dimensional composition appear to be more distant – however this isn’t always the case.

The general rule is the closer to the level of the horizon line, usually near the center of the frame or field of view, the father away the object appears. Of course this is dependent on a traditional composition – scenes with no horizon line (such as airplanes in flight) or objects that do not adhere to the ground plane would render this cue potentially invalid.


Relative Size
DepthCues_RelativeSizePerhaps the most obvious of all depth cues: If two objects are known to be the same size (like two adult velociraptors), but their absolute size is unknown, relative size cues can provide information about the relative depth of the two objects. Basically this cue comes down to the fact that the distance from an object to our eyeball is proportional to the object’s size divided by its retinal size.

We assume that in the two similar objects, the smaller objects (as it appears on the retina or camera recording device) appears to be more distant.


Shadowing & Shading
DepthCues_ShadowsIn most situations, objects appear in front of the contact shadows (ambient occlusion) they cast. Contact shadows are often the only source of a depth cue in a scene, giving the viewer the sole indication of the contact with the ground plane. Another example of this cue would be the shading of a crater, which is processed with the assumption that light naturally comes from above.

The way that light falls on an object and reflects off its surfaces, and the shadows that are cast by objects, provide an effective cue for the brain to determine the shape of objects and their position in space.


Motion Parallax
DepthCues_ParallaxIf the viewer or camera moves, the apparent relative motion of several stationary objects gives hints about their relative distance (this is actually due to the comparison of the velocity of the images across the retina).

Nearby things pass quickly, while far off objects appear stationary as they move at a much slower rate. When driving, consider the moon never seems to move at all or actually seems to be following you.


Textures and Details
DepthCues_DetailAs distance increases from the viewer, textured or patterned surfaces become more fine and eventually disappear completely. Texture gradients are only reliable depth cues when elements of similar size, shape, and spacing repeat in the scene.

There are normally three types of texture gradient cues that can be analyzed: Perspective (the width of texture elements decrease with distance), Compression (texture elements will shrink in height as the distance increases), and Density (the number of elements per unit area will increase with density).

For example, on a long gravel road, the gravel near the observer can be clearly seen of shape, size and color. In the distance, the road’s texture cannot be clearly differentiated.


Depth of Field / Defocus
DepthCues_DefocusSelective image defocus is commonly used in photography and film-making in order to establish the impression of depth. The depth of field of an optical system is the distance around the point of focus in which the image remains sharply focused.

The depth of field does not abruptly change from sharp to unsharp, but instead occurs as a gradual transition. In fact, everything immediately in front of or in back of the DOF convergence point begin to lose sharpness — even if not perceived by our eyes or the camera. Contrast decreases as defocus increases, contributing to the effect.

Because this article focuses on the pictorial aspects of depth cues, we’ve left out a few of the stereoscopic clues that would only exist in the real world; cues that would not exist on a two-dimensional image plane. These additional cues would include Accommodation (the brain processing the tension in the muscles that adjust the lens in the eye) and Convergence (the eyes moving slightly inward), and others.

Having a broad understanding of the types of depth cues will greatly increase the readability and instant viewer comprehension of your pictorial images – whether they are posters for print or animated footage.

Ciao!