Background tells: crowds, buildings and objects that do not make sense

Models focus on the subject. The background is where they cut corners.

The subject of a generated image gets the most attention, both from the model and from you. Move your eyes to the edges. People in a crowd may lack faces or merge into each other. Windows on a building may not line up floor by floor.

Objects in the background often have no function: a chair with three and a half legs, a railing that ends in mid-air, a door with no handle that leads nowhere.

Real backgrounds are full of boring, specific stuff, like cables, bins, parking signs and power outlets. Generated backgrounds are often vaguely tasteful and oddly empty.

Checklist

• Look at people in the background. Do they have faces, hands and feet?
• Follow straight lines like railings, tiles and window rows. Do they stay straight?
• Ask what each background object is for.
• Notice clutter. Real places are rarely tidy.

False alarms

Shallow depth of field blurs real backgrounds into shapes that look melted.

Test yourself

Put it into practice in the daily AI or Real game or the endless quiz.

Other tells

• Hands and fingers
• Text and signs
• Reflections and mirrors
• Light and shadows
• Skin and faces
• Physics and logic
• Repetition and symmetry
• Texture and noise