AI Image Prompt Glossary

Updated

This glossary defines the terms you meet in AI image and video prompts, in one or two sentences each. It covers prompt structure, camera and lighting vocabulary, and model-specific settings.

Prompt basics

  • Prompt: the text instruction given to an AI model to generate an image or a video.
  • Negative prompt: a list of things the model should avoid ("no text, no watermark").
  • Placeholder / variable: a bracketed word such as [PRODUCT] that you replace with your own subject before using a prompt.
  • JSON prompt: a prompt written as structured fields (subject, lighting, style) instead of a paragraph.
  • Reference image: an image uploaded with the prompt to keep a subject, a style or a layout.
  • Image-to-image: generating a new image from an existing one plus a prompt.
  • Image-to-video: animating a still image according to a prompt.

Format and settings

  • Aspect ratio: the shape of the image, written width:height (1:1 square, 4:5 vertical, 16:9 wide).
  • Seed: a number that fixes the random starting point, so the same prompt can reproduce a similar image.
  • Stylize: in Midjourney, how strongly the model applies its own aesthetic.
  • Style reference: an image whose look (colors, texture, mood) the model should imitate.
  • Character reference: an image of a person or character the model should keep consistent.
  • Upscale: increasing the resolution of a generated image.

Camera vocabulary

  • Focal length: the lens, in millimetres. Small numbers (24mm) are wide; large numbers (85mm, 135mm) are tight and blur the background.
  • Depth of field: how much of the image is in focus. Shallow means a sharp subject on a blurred background.
  • Bokeh: the soft blur of out-of-focus areas.
  • Close-up, medium shot, wide shot: how much of the subject and scene the frame includes.
  • Low angle / high angle: camera below or above the subject.
  • Top-down (flat lay): camera directly above, looking straight down.

Lighting vocabulary

  • Golden hour: warm, low sunlight shortly after sunrise or before sunset.
  • Softbox: a studio light that gives soft, even illumination.
  • Rim light: a light behind the subject that outlines its edges.
  • Rembrandt lighting: side light that leaves a triangle of light on the shadowed cheek.
  • Hard light / soft light: sharp-edged shadows versus gradual ones.
  • High key / low key: a bright image with few shadows versus a dark, contrasted one.

Style vocabulary

  • Photorealistic: looks like a photograph.
  • Editorial: the polished, art-directed look of magazine photography.
  • Cinematic: framed and color-graded like a film still.
  • Flat illustration: simple shapes and solid colors, no realistic shading.
  • Isometric: a 3D-looking view without perspective, common for dioramas and diagrams.
  • Character sheet: several views of the same character on one image, used as a reference.

Prompts to study

Real prompts from the network, each published with the result it produced:

More guides