

Turn the reference photo into a travel souvenir enamel pin badge. Compose it as a SCENE, not a single isolated object. Subject hierarchy: the defining landscape, terrain or landmark of the photo forms the main body of the badge and occupies most of its area. If a person appears prominently in the photo, keep them in the badge as a small, simplified figure at true relative scale within that landscape — the person is an accent, the landscape is the subject.Preserve the original spatial relationship and scale between the figure and the surroundings. How to render the person: flat enamel color blocks matching their real clothing and hair color from the photo. The face is a smooth plain area of light skin-tone enamel with no drawn facial features — do NOT render the person as a dark or black silhouette, and do NOT black out the face or head. Skin reads as a warm light enamel color, clearly lighter than the clothing.Styling: thin polished gold outline around the silhouette and along every internal divider, glossy enamel color fill, gentle even lighting with only a soft sheen on the gold lines, very subtle drop shadow. Outer contour follows the scene's own shape, not a plain rectangle. Background: flat dark navy coarse linen texture. Badge centered, filling about 60% of the frame.Avoid: black silhouette figure, blacked-out face, dark featureless head, portrait close-up, detailed facial features, person dominating the badge, cropping out the landscape, three-quarter angle, macro product photography, heavy specular glare, cartoon, realistic scene, text, watermark.
This image was generated with GPT Image 2 by OpenAI. The prompt on this page is the exact text used to create it.
The example was generated with GPT Image 2 using: 2:3 · Minimalist · Landscape. Reusing these settings gives the closest match.
Copy the prompt as-is first, then replace only the subject while keeping the style, lighting, and composition keywords. Generate a few variations — outputs differ between runs.
Yes. Prompts work best as starting points — adjust descriptors, aspect ratio, or style tags to make the output your own.