All categories/Image and video
Level Starter · Image Prompt Composition
Text to image: subject, light, lens, frame
'A cool cat' gives random results because it lacks executable visual information.
Prompt
Turn this into a usable image prompt.
What I want: {{idea}}
Use: {{use}} (cover / inline / poster)
Style reference: {{style}}
Size: {{size}}
Four blocks, all concrete:
1. Subject: who or what, doing what, wearing or made of what, expression or state.
2. Environment and light: time, weather, light direction, hard or soft, colour temperature.
3. Lens: shot size (close / medium / wide), angle (eye level / high / low), depth of field, motion blur or not.
4. Frame and composition: aspect ratio, where the subject sits, which side is left empty for text.
Then a shorter version under 20 words, plus 3 words that would push the image off target.What to state before it starts
Leave these unsaid and the model fills them with an assumption. Most 'wrong answers' start here.
- {{use}}: a cover needs empty space for the title; an inline image does not
- {{style}}: name a concrete reference, not 'premium feel'
- {{size}}: the ratio drives composition
Go through each of these
Where this kind of work most often goes wrong. Already in the prompt, so you do not have to remember it.
- The empty-space block is the most forgotten and the one that ruins covers.
- The off-target words list saves redraws.