Question
Should I describe the camera and lens in an image prompt?
Yes, when the framing matters. Terms such as close-up, wide-angle, low angle or shallow depth of field give the model a clear instruction about composition. Google's guidance for its image models recommends photographic and cinematic language to control the camera.
Lens numbers such as 85mm work as a style hint, not as a guarantee of real lens behavior, so describe the result you want as well: "background softly blurred behind a sharp subject".
Choose one viewpoint. Mixing "overhead" and "eye level" in one prompt leaves the model to pick.
In ImageQuarry you can ask for a change to the composition after the first image, such as a lower angle, and compare the new version with the old one.
| Term | What it usually changes |
|---|---|
| Close-up | The subject fills the frame |
| Wide-angle shot | More of the setting and stronger perspective |
| Low angle | The subject looks taller and more imposing |
| Shallow depth of field | A sharp subject against a blurred background |
| Overhead (flat lay) | The camera looks straight down, common for product layouts |
Source: Google's Gemini image generation best practices (control the camera); checked October 20, 2026.
Related questions
Last reviewed October 20, 2026 by ImageQuarry
Create an image. Edit it with words. Keep refining.
Frame the shot you want
Name the angle and distance, then adjust the framing in a follow-up request.
Start creatingAlready have an account? Sign in