AI image and video glossary: 50 terms every Higgsfield user should know

AI generation basics

• Prompt — the text instruction that describes what you want the model to create.
• Negative prompt — words describing what you do not want; some models support it, others prefer phrases like “no text”.
• Model — the AI system that generates the output (for example Kling, Seedance, Nano Banana, Soul).
• Text-to-image / text-to-video — generating from a text prompt only.
• Image-to-video — animating a still image; the image becomes the first frame.
• Start frame / end frame — the first (and optionally last) image a video must begin and end on.
• Reference image — an image given to the model to guide subject, style or composition.
• Seed — the random starting number of a generation; the same seed and settings can reproduce a similar result.
• Variation — another output from the same prompt, differing because of randomness.
• Inference / generation — the act of the model producing an output.
• Credits — the units Higgsfield uses to price each generation.
• Native audio — sound generated by the video model itself, in the same pass as the image.
• Multimodal — a model that accepts several input types: text, images, video, audio.
• Diffusion model — the family of models that create images by gradually removing noise.
• LoRA / fine-tuning — training a model on extra data (for example photos of a character) to specialise it.
• Character consistency — keeping the same face and look across generations. See our Soul guide.
• Hallucination / artefact — an error the model invents: extra fingers, melting objects, garbled text.

Editing and post-production

• Inpainting — regenerating only a selected area of an image.
• Outpainting — extending an image beyond its borders.
• Reframe — changing the aspect ratio of an image or video, often with outpainting.
• Upscaling — increasing resolution while adding plausible detail.
• Background removal — isolating the subject on a transparent background.
• Lip-sync — matching mouth movement to a voice track.
• Dubbing — replacing speech with a translation in another language.
• Motion transfer — applying the movement of one video to another subject. See Genjutsu.
• Colour grading — adjusting colours and contrast to give a film a unified look.
• Deflicker — reducing brightness flicker between video frames.

Framing and composition

• Aspect ratio — width-to-height ratio: 16:9 (landscape), 9:16 (vertical), 1:1 (square), 4:5 (portrait feed).
• Extreme wide / wide shot — shows the whole environment; the subject is small.
• Medium shot — subject from roughly the waist up.
• Close-up / extreme close-up — a face or a detail fills the frame.
• Low angle / high angle — camera below or above the subject; low feels powerful, high feels vulnerable.
• Over-the-shoulder — framed from behind one character, looking at another.
• Rule of thirds — placing key elements on imaginary lines dividing the frame in three.
• Depth of field — how much of the image is sharp; shallow depth of field blurs the background.

Camera movement

• Dolly in / out — the camera physically moves toward or away from the subject.
• Truck left / right — the camera slides sideways.
• Pan / tilt — the camera rotates horizontally / vertically from a fixed point.
• Orbit — the camera circles around the subject.
• Crane / jib — the camera rises or descends smoothly.
• Handheld — natural, slightly shaky movement as if held by a person.
• Zoom — changing focal length without moving the camera; feels different from a dolly.
• Static shot — the camera does not move.

Lenses and light

• Focal length — 24mm (wide), 35–50mm (natural), 85mm+ (portrait, telephoto).
• Anamorphic lens — a cinema lens giving wide frames, oval bokeh and horizontal flares.
• Macro — extreme close-up photography of small details.
• Bokeh — the soft, blurred look of out-of-focus lights.
• Golden hour / blue hour — warm light just after sunrise or before sunset / cool light just after sunset.
• Key light, fill light, rim light — the main light, the light that softens shadows, and the light that outlines the subject from behind.
• Practical light — a light source visible in the scene (lamp, neon sign).
• Volumetric light — visible beams of light through fog, dust or haze.
• Film grain — the fine texture of analogue film, often added for a cinematic feel.

Frequently asked questions

Do I need to learn all these terms?

No. Start with framing (wide, close-up), one or two camera moves (dolly in, orbit) and light (golden hour, rim light). They already improve most prompts.

What is the difference between a zoom and a dolly?

A zoom magnifies the image from the same position; a dolly moves the camera through space, which changes perspective and feels more cinematic.

Facts checked: October 29, 2026. Features, models and plans change often — always confirm on the official site.