Depth and perspective

A photograph or film frame is flat, but the world it shows is not. Much of the craft of composition lies in creating a convincing sense of depth, drawing the viewer into the scene rather than presenting a flat arrangement of shapes. This lesson covers the visual cues the brain uses to perceive depth, linear perspective and vanishing points, how camera position and focal length shape perspective, depth of field as a tool for separating planes, and layering foreground, middle ground and background.

Depth cues

With two eyes, people judge depth partly from the slightly different view each eye receives (binocular disparity). Flat images cannot use this cue, so they rely on monocular cues that work with one eye, which painters discovered over centuries:

• Overlap (occlusion): an object that covers part of another is in front of it. The strongest single depth cue.
• Relative size: identical objects look smaller when farther away, so a row of trees or people shrinking into the distance creates depth.
• Linear perspective: parallel lines in the world, such as rails, roads or building edges, converge toward a vanishing point.
• Height in the frame: objects on the ground that are farther away appear higher in the frame, closer to the horizon.
• Atmospheric (aerial) perspective: distant objects look paler, lower in contrast and bluer because of haze. Fog and smoke exaggerate it.
• Texture gradient: surfaces such as grass, cobblestones or waves show finer texture with distance.
• Light and shadow: shading reveals form, and cast shadows place objects on surfaces.
• Focus: limited depth of field blurs planes in front of and behind the subject, separating them.
• Motion parallax (in moving images): nearer objects move across the frame faster than distant ones as the camera moves.

Size and distance

The size of an object in an image is inversely proportional to its distance from the camera. For an object of height \(h\) at distance \(d\), photographed with a lens of focal length \(f\), the height of its image on the sensor is approximately:

This simple relation explains why foreground objects can dominate a frame: a person 2 m from the camera appears five times as tall as an identical person 10 m away.

Linear perspective

Linear perspective, formalized by Renaissance artists such as Filippo Brunelleschi and Leon Battista Alberti in the fifteenth century, describes how parallel lines in the world appear to converge. All lines parallel to each other (and not parallel to the image plane) meet at a vanishing point. Lines parallel to the ground meet at vanishing points on the horizon line, which is always at the height of the camera (the viewer's eye level).

• One-point perspective: when the camera faces a set of parallel lines head-on, such as looking straight down a corridor, road or railway, they converge to a single vanishing point, often near the center. It creates strong, symmetrical depth.
• Two-point perspective: when the camera faces the corner of a building, its horizontal edges converge to two vanishing points on the horizon, left and right, while vertical edges stay vertical (if the camera is level).
• Three-point perspective: when the camera tilts up or down, vertical lines also converge, to a third vanishing point above or below. Looking up at a tall building makes it taper upward ("converging verticals"), which can look dramatic or simply wrong depending on intent.

Camera position, focal length and perspective

A common misconception is that wide-angle lenses "distort" perspective and telephoto lenses "compress" it. Perspective, the relative sizes and positions of objects, depends only on where the camera is. Focal length determines only how much of the scene is included (the field of view). What changes in practice is that photographers move closer with wide lenses and farther away with long ones, and that change of distance changes perspective.

• Close with a wide lens: nearby objects become very large relative to distant ones, exaggerating depth and distances. Faces photographed close up look distorted, with large noses, which is why portraits are usually taken from farther away with longer lenses.
• Far with a long lens: the ratio of distances between near and far objects becomes small, so they look similar in size and close together, "compressing" the scene. Long lenses make background mountains loom behind a subject and crowds look packed.

Field of view is set by focal length and sensor size. For a sensor of width \(w\) and focal length \(f\):

A 50 mm lens on a full-frame sensor (36 mm wide) has a horizontal field of view of about 40°, similar to the central attention area of human vision, which is why it is called a "normal" lens. Smaller sensors give a narrower view with the same lens; their "crop factor" (1.5 or 1.6 for APS-C, 2 for Micro Four Thirds) multiplies the focal length to give the full-frame equivalent field of view.

Depth of field

Only one distance is in perfect focus, but a range in front of and behind it, the depth of field (DOF), appears acceptably sharp. Depth of field increases with smaller apertures (higher f-numbers), shorter focal lengths and greater subject distance, and is roughly:

where \(N\) is the f-number, \(c\) the acceptable blur (circle of confusion, about 0.03 mm for full frame), \(d\) the subject distance and \(f\) the focal length. Shallow depth of field isolates a subject from a busy background, creating a clear figure-ground relationship (Lesson 1) and directing attention; deep depth of field keeps everything sharp, suiting landscapes, architecture and scenes where the environment matters. The hyperfocal distance \(H \approx f^2/(Nc)\) is the focus distance that keeps everything from \(H/2\) to infinity acceptably sharp, a technique used in landscape photography.

Layering: foreground, middle ground and background

The most reliable way to create depth is to arrange the scene in layers. A foreground element (rocks, flowers, a doorframe, a person's shoulder), a subject in the middle ground and a background behind it give the eye a path from near to far. Each layer should be distinct in tone, focus or size so the planes separate. In landscapes, a strong foreground anchors the composition and gives scale; in interiors and film scenes, staging action in depth, with characters at different distances, adds richness and allows a single shot to show relationships that would otherwise need several cuts. Cinematographers call this deep staging, and directors such as Orson Welles and Gregg Toland in Citizen Kane (1941) famously used deep focus to keep near and far planes sharp at once.

Lighting reinforces depth: alternating light and dark planes (a dark foreground, a lit subject, a darker background) separate layers even when they are similar in color. A rim or back light outlines a subject against the background, and haze or smoke makes distant layers recede.

Lesson 8 shows how light and shadow model form and create depth, and Lesson 11 how camera movement reveals space.