The camera words a beginner needs for AI video fit into four small groups: how much is in the frame (shot size), where the camera sits (angle), how the camera moves, and what the lens does. About twenty words cover all four.
Pick one word from each group for each clip, write it plainly in your prompt, and the video tool knows where to put its camera instead of guessing.
- Shot size says how close we are: extreme close-up, close-up, medium shot, wide shot, over-the-shoulder.
- Angle says where the camera sits: eye level, low angle, high angle, bird’s-eye view, Dutch angle.
- Move says what the camera does: static, pan, tilt, dolly in, tracking, orbit, handheld.
- Lens says how the picture feels: wide-angle, telephoto, and shallow depth of field for a blurred background.
- One of each, no more. One size, one angle, one move per clip. Stack five moves and the clip tends to muddle them.
An AI video tool is like a camera operator who has watched every film ever made but cannot read your mind. “Make it dramatic” leaves them guessing. “Low angle, slow dolly in” tells them exactly where to stand and what to do. Camera words are the short, shared language that turns a mood into an instruction.
Why do camera words work in AI video prompts?
The plain point: the video tools ask for them. Their own prompt guides list the camera as one part of a good video prompt, next to the subject and the action.
Google’s guide for its Veo video model lists camera position and movement, shot framing, and focus and lens effects among the things to put in a prompt.1 Runway’s structure for a text-to-video prompt even starts with the camera.2
The words are old film words, not new AI jargon. That helps you: once you know them, they work in any tool, and in any film class too. This page keeps to the beginner set. The full vocabulary, with more than twenty more terms, is on camera grammar.
Shot sizes: how much is in the frame
The plain point: shot size is how close the viewer feels. Close shots show feeling and detail; wide shots show the place. The meanings below are from Google’s Veo prompt guide; the lines to paste are this site’s own examples.
| Word | What it looks like | Line to paste |
|---|---|---|
| Extreme close-up | Looks like: one small detail fills the screen, such as an eye or a drop of water.3 | Paste: extreme close-up of her eyes as they open |
| Close-up | Looks like: the face fills the frame, so you see every expression.3 | Paste: close-up of an old man’s face as he smiles |
| Medium shot | Looks like: a person from about the waist up, with a little of the room around them.3 | Paste: medium shot of a baker kneading dough at the counter |
| Wide shot | Looks like: the person is small and the place is big. Often used as the first shot, to show where we are.3 | Paste: wide shot of a lone hiker crossing a snowy valley |
| Over-the-shoulder | Looks like: we look past one person’s shoulder at someone or something else. Common in conversations.3 | Paste: over-the-shoulder shot from behind the customer, looking at the shopkeeper |
Close for feeling, wide for place. If you name only one camera word, make it the shot size.
Camera angles: where the camera sits
The plain point: the angle is the height of the camera compared with the subject’s eyes. It changes how strong or small the subject looks, which is why camera grammar calls it the most reliable lever you have.
| Word | What it looks like | Line to paste |
|---|---|---|
| Eye level | Looks like: a normal, neutral view, as if you were standing there.3 | Paste: eye-level shot of two friends talking on a park bench |
| Low angle | Looks like: the camera is low, looking up, so the subject looks powerful.3 | Paste: low-angle shot looking up at a firefighter in front of the truck |
| High angle | Looks like: the camera is above, looking down, so the subject looks small.3 | Paste: high-angle shot of a child alone in a big empty classroom |
| Bird’s-eye view | Looks like: straight down from above, like a map. Also called top-down or overhead.3 | Paste: bird’s-eye view of a busy crossing, people moving in every direction |
| Dutch angle | Looks like: the camera is tipped sideways, so the horizon slopes and the scene feels uneasy.3 | Paste: Dutch angle on a man in a dim corridor, the horizon tilted |
Low makes strong, high makes small, eye level stays neutral. Choose the angle for how you want the viewer to feel about the subject.
Camera moves: what the camera does during the clip
The plain point: a move is what the camera does while the clip plays. Pick one, and add a speed word such as “slow”. Google’s own examples use phrases like “slow pan left” and “slow zoom in”.4
| Word | What it looks like | Line to paste |
|---|---|---|
| Static (locked-off) | Looks like: the camera does not move at all; only the subject does.3 | Paste: static shot, the camera completely still, only the candle flame moving |
| Pan | Looks like: the camera stays in one spot and turns left or right, like turning your head.3 | Paste: slow pan left across the harbour at sunrise |
| Tilt | Looks like: the camera stays in one spot and tips up or down, like nodding.3 | Paste: slow tilt up from the boots to the face of the climber |
| Dolly in / dolly out | Looks like: the whole camera travels toward the subject (dolly in, also called push-in) or away from it (dolly out, or pull back).3 | Paste: slow dolly in toward the woman at the window |
| Tracking shot | Looks like: the camera travels alongside someone who is moving, keeping pace with them.5 | Paste: tracking shot beside a runner on a beach path at dawn |
| Orbit | Looks like: the camera circles around the subject. A part-circle is called an arc.6 | Paste: the camera slowly orbits a bronze statue in a quiet square |
| Handheld | Looks like: a person is holding the camera, so it moves and shakes a little. It feels real and close.3 | Paste: handheld camera following a cook through a busy kitchen, slight natural shake |
Zoom is not a dolly. In a zoom the camera stays put and the lens magnifies; in a dolly the camera itself travels.3 In this site’s reading, a dolly feels like walking into the room, and a zoom feels like someone pointing: “look at that”. If you mean one, do not write the other.
One move, plus a speed word. “Slow dolly in” is a clear instruction; “dolly in, pan and orbit” asks the tool to compromise.
Two lenses and one focus word
The plain point: the lens decides how much of the world fits in the frame and how near or far the background feels. A beginner needs two lens words and one focus word.
- Wide-angle lens. Fits a lot in, and makes near things look big and far things small. Use it for big places and drama.3 Paste: wide-angle lens, the cathedral ceiling soaring above tiny figures.
- Telephoto lens. Fits less in, and pulls the background up close behind the subject. Use it to pick one person out of a crowd.7 Paste: telephoto lens, the cyclist framed tight, the city skyline stacked close behind her.
- Shallow depth of field. Only a thin slice of the picture is sharp, and the rest melts into blur. It points the eye at one thing.3 Paste: shallow depth of field, only her face in focus, the party behind a soft blur.
Lens numbers in millimetres, such as 24mm, 50mm or 85mm, do the same job in more detail. In Nikon’s words, “The longer the focal length, the narrower the angle of view and the higher the magnification.”8 The lens table on camera grammar gives five common numbers with lines to paste.
How to put camera words into a prompt
The plain point: one shot size, one angle, one move, and a lens only if you care. Put them in plain words, near the start or wherever your tool’s guide puts the camera.
One of each. In this site’s use, stacking several moves in one short clip tends to give a muddled average of them. Runway is a partial exception: for its Gen-4.5 model it says combining camera terms is encouraged for cinematic sequences.9 So start with one, and add a second only once the first works.
A still camera needs saying plainly. Runway’s own static-shot example spells it out in a full sentence.10
Moves need time. In this site’s reading, a slow move needs a clip long enough to show it; in a very short clip, a big move has no room and is better swapped for a static shot.
Check for a camera setting. Some tools also have camera controls you set with a button, not words. Kling documents a camera movement control, and Higgsfield a library of named camera presets.11 If your tool has one, set it and name the same move in the prompt, so the two agree.
Size, angle, move, then a lens if it matters. Write each one once, in plain words, and keep the camera instruction separate from what the subject does.
Questions people ask
What is the most important camera word to start with?
Shot size: it decides most of what ends up on screen. Then the angle. This is this site’s own advice for beginners; camera grammar explains why angle and composition are understood more reliably than movement.
Do camera words work in every AI video tool?
The major video tools’ guides use them: Google lists terms such as eye-level, dolly shot, wide shot and close-up for Veo, and Runway’s text-to-video structure starts with “[Camera] shot of” (sources above). Each tool orders a prompt differently, so check its cheat sheet.
Why does the camera move when I asked for a still shot?
Runway’s own static-shot example spells it out in a full sentence (source above). Write that the camera is entirely still and that only the subject moves, and avoid the word “no” on tools that do not read negatives; fixing flicker and morphing covers more unwanted motion.
Where to go next
- “I want the full vocabulary.” Read camera grammar: twenty-four shot types, fourteen moves, speed words and lenses.
- “Now help me write the whole prompt.” Follow how to write your first AI video prompt, or let the prompt generator draft one.
- “What about light?” See colour, light and mood.
- “I want an exact ending.” Use first frame and last frame.
- “What do these words mean?” See text-to-video, image-to-video and keyframe.
- “Take me through it step by step.” Follow the path make your first AI video.
- Related: Higgsfield Camera Presets.
QUICK CHECK · 3 QUESTIONS
Pick an answer to see if you got it. Nothing is recorded.
1. Which four groups does the page sort beginner camera words into?
- Colour, sound, speed and length
- Shot size, angle, camera move and lens
- Subject, action, setting and style
Show the answer
Answer: Shot size, angle, camera move and lens. The page says the camera words a beginner needs fit into four groups: shot size, angle, camera move and lens.
2. What does a low angle do to the subject, according to the page?
- Makes the subject look powerful
- Makes the subject look small and vulnerable
- Shows the scene like a map from above
Show the answer
Answer: Makes the subject look powerful. The page says a low angle puts the camera below the subject looking up, so the subject looks powerful.
3. What is the difference between a dolly and a zoom?
- A dolly tips the camera up; a zoom turns it sideways
- There is no difference; they are two names for one move
- In a dolly the camera travels; in a zoom the lens magnifies and the camera stays put
Show the answer
Answer: In a dolly the camera travels; in a zoom the lens magnifies and the camera stays put. The page explains that a dolly moves the whole camera, while a zoom changes the lens and the camera does not move.
- 1Google AI for Developers, Generate videos with Veo 3.1 in Gemini API, read at source 13 Oct 2026camera positioning and motion, “Control the camera’s location and movement using terms like” “aerial view, eye-level, top-down shot, dolly shot, or worms eye”; composition, “How the shot is framed, such as” “wide shot, close-up, single-shot or two-shot.” ↑
- 2Runway Academy, Prompting Guide, undated, read at source 13 Oct 2026“[Camera] shot of [a subject/object] [action] in [environment]. [Supporting component descriptions]” ↑
- 3Google Cloud, Video generation prompt guide, last updated 9 Oct 2026, read at source 13 Oct 2026extreme close-up “isolates a very small detail of the subject, such as an eye or a drop of water.” ↑aclose-up “frames the subject tightly, typically focusing on the face to emphasize emotions or a specific detail.” ↑bmedium shot “shows the subject from approximately the waist up, balancing detail with some environmental context.” ↑cwide shot “shows the subject within their broad environment, often used to establish location and context”. ↑dover-the-shoulder “frames the shot from behind one person, looking over their shoulder at another person or object, common in conversations.” ↑eeye-level “offers a neutral, common perspective, as if viewed from human height.” ↑flow-angle “positions the camera below the subject, looking up, making the subject appear powerful or imposing.” ↑ghigh-angle “places the camera above the subject, looking down, which can make the subject seem small, vulnerable”. ↑hbird’s-eye view “a shot taken directly from above, offering a map-like perspective of the scene.” ↑iDutch angle, “the camera is tilted to one side, creating a skewed horizon line, often used to convey unease”. ↑jstatic, “the camera remains completely still, there is no movement.” ↑kpan, “the camera rotates horizontally left or right from a fixed position.” ↑ltilt, “the camera rotates vertically up or down from a fixed position.” ↑mdolly, “the camera physically moves closer to the subject or further away.” ↑nhandheld, “the camera is held by the operator, resulting in less stable, often jerky movements”. ↑ozoom, “the camera’s lens changes its focal length to magnify or de-magnify the subject.” ↑pwide-angle lens “captures a broader field of view than a standard lens.” ↑qshallow depth of field, “an optical effect where only a narrow plane of the image is in sharp focus”. ↑r
- 4Google Cloud, Video generation prompt guide, last updated 9 Oct 2026, read at source 12 Oct 2026 (as cited on this site’s camera grammar page): its examples include “slow pan left” and “slow zoom in”.↑
- 5Runway, Camera Terms, Prompts, & Examples, read at source 12 Oct 2026 (as cited on this site’s camera grammar page): tracking, “Camera follows alongside a moving subject”.↑
- 6Runway, Camera Terms, Prompts, & Examples, read at source 12 Oct 2026 (as cited on this site’s camera grammar page): orbit, “Camera circles completely around the subject”. Google Cloud, Video generation prompt guide, last updated 9 Oct 2026, read at source 13 Oct 2026arc, “the camera moves in a circular or semi-circular path around the subject.” ↑
- 7Google Cloud, Video generation prompt guide, last updated 9 Oct 2026, read at source 12 Oct 2026 (as cited on this site’s camera grammar page): telephoto, “narrows the field of view and compresses perspective, making distant subjects appear closer”.↑
- 8Nikon, Understanding Focal Length, updated Dec 2025, read at source 12 Oct 2026 (as cited on this site’s camera grammar page): “The longer the focal length, the narrower the angle of view and the higher the magnification.”↑
- 9Runway, Camera Terms, Prompts, & Examples, read at source 12 Oct 2026 (as cited on this site’s camera grammar page): “Yes—combining terms is encouraged, especially for cinematic sequences.”↑
- 10Runway, Camera Terms, Prompts, & Examples, read at source 12 Oct 2026 (as cited on this site’s camera grammar page): “The camera is entirely motionless for the duration of the scene, with movement only occurring from the building moving within the locked perspective.” (wording re-read at source 12 Nov 2026)↑
- 11Kling AI, Kling AI Camera Control Guide, 24 Nov 2025, read at source 12 Oct 2026 (as cited on this site’s camera grammar page): “six basic camera movements, including ‘horizontal, vertical, zoom, pan, tilt, roll’”. Higgsfield, Higgsfield Camera Controls, read at source 12 Oct 2026 (as cited on this site’s camera grammar page), page title: “Higgsfield Camera Controls – 50+ Cinematic AI-Motion Presets”.↑
Read at source 13 Oct 2026: Google Cloud, Video generation prompt guide (last updated 9 Oct 2026); Google AI for Developers, Generate videos with Veo 3.1 in Gemini API; Runway Academy, Prompting Guide.
Reused from this site’s camera grammar page, with the read date recorded there (12 Oct 2026): Runway’s Camera Terms, Prompts, & Examples; Nikon’s Understanding Focal Length; Kling’s Camera Control Guide; Higgsfield’s Camera Controls; and Google’s speed-word and telephoto examples.
The lines to paste and the full prompts are this site’s own examples, not claims from these sources. The camera-operator comparison and the advice marked “this site’s reading” or “this site’s use” are this site’s own.
Free printable AI cheat sheets, and AI news every week.
Sixteen one-page AI guides and checklists in one free PDF, no sign-up. And if you want to keep up, the newsletter covers what changed in AI each week, in plain English, with the sources.
SUBSCRIBE FREE ↗ GET THE CHEAT SHEET PACK