Where do I put the camera words in my prompt?
After the sentence that says who is in the frame and what they do. Name the shot size first, then the angle, the lens and, for video, the move. The example prompt shows the order.
Get camera angle prompt words for your AI persona: pick a shot size, an angle, a phone lens and, for video, a camera move, each shown as a small diagram. It is for anyone writing image or video prompts for an AI influencer, and it gives the camera words, a JSON camera block and a full example prompt for a vertical photo or clip.
Free · No signup · Runs in your browser
Empty fields use the example text in grey.
Choose Photo or Video. Click a shot size, an angle and a lens, and for video a camera move.
Add your persona, what they’re doing and where, or keep the example text.
Copy the camera words into your own prompt, or copy the full example prompt or the JSON block.
A vertical frame is tall and narrow, so it holds less of the place and more of the person. Pick the size by what the viewer needs to see in that clip.
| Shot size | What the frame holds | Use it for |
|---|---|---|
| Extreme close-up | The eyes, or one small detail | A moment of emotion, a texture, a product detail |
| Close-up | Head and the top of the shoulders | The hook, a reaction, a line said straight to camera |
| Medium close-up | Head to mid-chest | Talking to camera, the everyday phone frame |
| Medium shot | Head to hips | Talking while the hands show something |
| Medium full | Head to below the knees | Moving, cooking, working, with some of the place |
| Full shot | Head to feet | Outfits, walking, a whole movement |
| Wide and extreme wide | The place, with the person small in it | An opener, a change of place, a mood shot |
Cutting between two sizes, such as a close-up and a medium shot, keeps a talking video moving without changing the place.
On a phone, the 1x lens gives the look people expect from a phone. The 2x lens keeps faces in natural proportion; the ultra wide stretches whatever is close to it.
Give each clip of your AI persona its own shot size, angle and camera move: an image prompt first, then a video prompt, on your own API keys. How making videos works →
After the sentence that says who is in the frame and what they do. Name the shot size first, then the angle, the lens and, for video, the move. The example prompt shows the order.
Eye level, straight on, in a medium close-up or a medium shot on the 1x or 2x lens, as if the phone sits on a shelf. Cut to a close-up for the lines that matter most.
One. A short clip gives the camera a few seconds, so name one move in plain words, such as a slow push-in. Keep the next clip still or give it a different move.
Plain film words: a static or locked-off shot, a slow push-in, a natural phone zoom, one continuous shot. Google’s Gemini Omni prompt guide uses these words. Name one move per clip.
A shot with the camera tilted to one side, so the horizon runs on a slant. It feels energetic or uneasy. Use it once in a video, not in every clip.
A close-up holds the head and the top of the shoulders. A medium close-up goes down to the middle of the chest, which leaves room for small hand moves while talking.