The Story in the Frame

Composition, light and subject decide whether a picture talks to the person looking at it. A practical guide to the rule of thirds, composition types, balance, point of view, depth of field, framing and negative space.

Photography · Composition · 10 min

A picture is a message sent to someone you will probably never meet. They will give it a second or two, and in that time the image has to tell them where to look, what is happening and how to feel about it. That is why the most important quality of a good image is not sharpness or resolution but its **ability to communicate** with the viewer. A good image tells a story, and it does that with three tools: **composition**, which arranges the frame; **lighting**, which sets the mood and decides what can be seen; and above all the **subject**, the person or thing the story is about. Fig. 2 holds the road still and lets you change those three tools. A figure under hard noon light reads as a plain fact, the same figure at golden hour reads like a memory, and under a single lamp at night it reads like a question. Take the traveller away and the picture stops being about anyone.

Of the three, the **subject matters most**, because it is what the viewer's eye searches for first. Eye-tracking studies going back to Alfred Yarbus in the 1960s show that people look at faces and figures first, and that the eye is pulled by contrast: a bright patch on dark, a saturated colour among muted ones, a sharp edge in a soft field. Fig. 3 runs a simple contrast-based attention model over our frame and paints a heat map of where it predicts the eye will go. Give the traveller a red coat and they draw far more of the predicted attention than their size deserves. Dress them in grey and the tree, the fence and the bright sky start to compete. A good composition makes sure the strongest pull in the frame is the thing the story is about.

Two lines across, two lines down

The **rule of thirds** is the first composition guideline most people learn, and its grid is built into nearly every camera and phone. Divide the frame into three equal columns and three equal rows. The four lines are good places for strong edges such as horizons and tree trunks, and the four crossing points, often called **power points**, are good places for the subject or for the subject's eyes. A subject placed there leaves the rest of the frame free to show context, and it creates a slight imbalance that keeps the eye moving instead of settling in the dead centre. Drag the traveller around Fig. 4 (the camera pans and tilts to follow, as yours would) and watch the readout as their eyes land on a power point.

The same idea applies to the **horizon**. A horizon across the middle cuts the picture into two equal halves that compete with each other. Move it to a third and you choose what the picture is about. On the lower third the sky takes two thirds of the frame, so the weather, the light and the open space become the story. On the upper third the ground wins: the road, the distance still to walk, the grass at the traveller's feet. Fig. 5 measures the split as you move the line.

The rule of thirds is a simplified form of older proportions. Painters, and later photographers, also used the **golden ratio**, about 1 : 1.618, which puts its lines at 38.2% and 61.8% of the frame instead of 33.3% and 66.7%. The difference is small, as Fig. 6 shows, and both work for the same reason: they move the subject off centre by a comfortable amount. Treat them as starting points, not laws. A portrait that wants confrontation, a symmetrical building or a subject staring straight into the lens often works better in the centre, and breaking the rule on purpose is part of using it well.

Shapes the eye can follow

Beyond the grid, photographers describe compositions by the shape they make. **Centred and symmetrical** frames feel formal and still. **Leading lines** pull the eye into the picture. **Diagonals** add energy, **triangles** add stability, a **frame within the frame** focuses attention, the **golden spiral** guides the eye along a curve, and **negative space** isolates the subject. Fig. 7 is a contact sheet of the same road and traveller composed eight ways. None of them is the correct one; each tells the viewer something different about the same moment.

**Leading lines** are any lines the eye can travel along: roads, fences, rivers, railings, shadows, even a queue of people. Parallel lines converge toward a vanishing point in perspective, so they act like arrows. When that point sits on the subject, the whole image points at them, which is why a low camera on a long road is such a common opening shot in films. In Fig. 8 the moving dots trace the paths a viewer's eye can follow. Move the traveller off the road and the lines still run to the horizon, now leading the eye away from the story.

**Symmetry** is the opposite strategy. When the left half of a frame mirrors the right, the eye reads the image as calm and ordered, sometimes uncanny. Directors such as Stanley Kubrick and Wes Anderson built whole visual styles on it. A centred subject in a symmetrical frame feels deliberate, as if the world has arranged itself around them. Fig. 9 measures how closely the two halves of the picture mirror each other. Move the camera and the subject a few metres to one side and the score drops, and the feeling of order goes with it.

**Diagonals and triangles** are about energy and stability. A road crossing the frame from corner to corner feels like movement even in a still image, because a diagonal looks like something that could fall or slide. A triangle, such as two people with a tree rising behind them, feels grounded: it has a wide base and a single peak that holds the group together. Horizontal lines feel calm and at rest. Fig. 10 draws the shape each version makes.

Weighing the frame

Every element in a frame has **visual weight**. Large things weigh more than small ones, bright and saturated things more than dull ones, sharp things more than blurred ones, and people more than objects. **Balance** is how that weight is spread around the centre, and it works like a seesaw: a small element far from the centre can balance a large one close to it. Fig. 11 measures balance by running the attention model from Fig. 3 and multiplying each pixel's pull by its distance from the centre line. Move the traveller, move the tree and change the coat, and watch the seesaw.

**Symmetrical balance** puts equal weight on each side and feels formal and stable. **Asymmetrical balance** pairs unequal elements, such as a big tree on one side and a small figure on the other, and feels more natural and alive. Fig. 12 measures four arrangements. A big tree alone tips the frame hard to one side. A small grey figure on the far side helps only a little; the same figure in red pulls the frame much closer to level. Colour and contrast can do the work that size cannot. An unbalanced frame is not wrong, but it should be unbalanced on purpose, for example to make the viewer feel uneasy.

Where you stand changes who they are

**Point of view** is where the camera stands, and it changes how we judge the subject. From a **worm's-eye view** the traveller towers over us against the sky and looks powerful, even threatening. At **eye level** they are our equal. From **above** they shrink against the ground and look small, lost or vulnerable, and from high overhead the picture starts to become a map. Fig. 13 raises the camera from the ground to twenty metres while keeping it aimed at the traveller, who stays eight metres away.

**Perspective** also depends on distance. A wide lens used up close exaggerates depth: near things look huge and the background falls far away. A long lens used from far back **compresses** depth, so the background seems to press in behind the subject. In Fig. 14 the camera backs away as the focal length grows, so the traveller stays the same size and only the background changes. The tree forty metres behind them grows from less than half their height to well over twice it. Zooming alone does not cause this; the camera's distance does, and the focal length lets you keep the framing.

Choosing what is sharp

**Depth of field** is the zone in front of and behind the point of focus that looks acceptably sharp. A **shallow** depth of field isolates the subject against a soft background and tells the viewer exactly where to look. A **deep** one keeps the whole scene readable, so the setting becomes part of the story. Fig. 15 is a working depth-of-field calculator for an 85mm lens on a full-frame camera, using the standard 0.03 mm circle of confusion. The blur in the picture is computed from the same optics for every object at its real distance, so you can see what f/1.4 and f/16 actually look like.

Three things control it: the **aperture**, the **focal length** and the **distance** to the subject. A wider aperture (a smaller f-number), a longer lens and a closer subject all make the sharp zone thinner. The effect is far from linear. In Fig. 16, with the subject five metres away, an 85mm lens at f/1.4 keeps about 30 centimetres sharp, while a 35mm lens at f/16 keeps everything sharp from 1.7 metres to infinity. Sensor size plays a part too, which is why phones, with their tiny sensors, use software to fake the background blur of a large camera.

A frame inside the frame

**Framing** means using something in the scene, such as a doorway, a window, an arch or overhanging branches, to draw a second border around the subject. It works as a pointer, because the eye follows the inner frame to whatever sits inside it. It adds depth by putting a near layer in front of a far one. And it can carry part of the story: a doorway suggests we are watching from inside a house the traveller has just left, a window suggests distance or longing, branches suggest someone hidden in the trees, an arch makes the road a destination. Switch frames in Fig. 17 and notice how each changes who seems to be watching.

A frame does not have to be sharp. Many cinematographers let the foreground fall soft, so it reads as shape and shadow rather than detail and keeps attention on the subject. Focus on the frame instead and the subject becomes the blurred mystery in the distance, which is a different story altogether. Fig. 18 renders the doorway with three lens settings so you can compare the layers.

The power of empty

**Negative space** is the empty area around the subject: sky, fog, a plain wall, still water. It is not wasted space. It gives the subject room, makes it stand out clearly, and can say something on its own: loneliness, freedom, silence, scale. Minimalist photographers often let the subject fill only a tiny part of the frame. Fig. 19 measures exactly how much of the frame the traveller occupies as they walk away into the fog.

Negative space also has a direction. When a subject looks or moves toward one side of the frame, viewers expect space on that side for them to look or move into, often called **lead room** or looking room. With the space in front of them the traveller seems free to go on. Turn them toward the edge, with all the space behind, and they seem blocked, or haunted by something they have left behind. Fig. 20 shows the three versions. The second one breaks the rule on purpose, and it works only when the story wants that tension.

Composing in camera and on the computer

Most tools help you apply these ideas. Camera and phone camera apps can show a rule-of-thirds grid on screen while you shoot. When you edit, cropping lets you recompose: **Adobe Lightroom Classic** shows overlays while you crop (grid, thirds, diagonal, triangle, golden ratio, golden spiral and aspect ratios; press O to cycle through them), and **Photoshop**'s Crop tool has a similar overlay menu. Fig. 21 is a simple crop tool modelled on those overlays. Drag the crop box, change its shape and cycle the overlays to find a stronger frame inside a loose one.

One sentence per picture

A useful habit is to describe the picture in one sentence before you take it. Each choice in this post adds a clause to that sentence: where the subject stands, where the camera stands, how much is sharp, what frames them and what light falls on them. Fig. 22 builds a frame from five choices and writes the sentence your frame says. If the sentence is not the story you meant, change a choice.

Here is everything on one page: what each tool does to the story, and when it pays to break it. The thumbnails are rendered from the same scene.

Rules like these are useful because they describe how people read pictures, not because a picture has to obey them. The strongest images usually follow some of them and break others for a reason, and every choice serves the subject and the story. So before the next shot, look at the empty frame and decide what it should say. Fig. 24 is that frame: the road at night, the lamp lit, the power point waiting.