Stubborn ‘I Can’t Hear You’ 3D Caricature Figures

Turn every subject in a reference image into a cute, stubborn 3D caricature figure caught refusing to listen. The prompt preserves each subject’s distinctive features while specifying a comic pose, pastel studio lighting, and a pink embroidered outfit.

Cheolyon Republished on Nov 13 0 Bookmarks 0 Copies
Bookmark 0

Example outputs

AI-generated image of Cute, stubborn 2–2.5-head-tall 3D caricature figures cover their ears and stomp in a gray pastel studio, with pink rim lighting and a detailed embroidered outfit.
AI-generated image of Cute, stubborn 2–2.5-head-tall 3D caricature figures cover their ears and stomp in a gray pastel studio, with pink rim lighting and a detailed embroidered outfit.

Prompt template

Format
Text
Inputs
Reference_image (image) · Required
Outputs
Rendered_scene (image)
Reinterpret every subject in the uploaded reference image as one of the world’s most stubborn yet lovable 2–2.5-head-tall 3D caricature figures. ABSOLUTE IDENTITY LOCK 300%. Preserve each subject’s original face shape, eye shape, irises, nose, lips, hair, skin tone, age impression, and distinctive features exactly. Keep every original subject; do not add, remove, or duplicate anyone. This prompt applies to people, characters, Blythe dolls, BJD dolls, and animals. For animals, preserve species, fur markings, ears, tail, and anatomy. For dolls, preserve facial sculpt, joints, and materials. Do not make adults look younger.

Give the figure voluminous twin-bun pigtails tied with pink hair elastics, with wispy strands sticking out, and a pastel pink pinafore-style bustier dress embroidered with flowers. Prioritize preserving each subject’s distinctive original hair and clothing. Give the figures oversized heads, plump cheeks, and short, delicate limbs. Show them squeezing their eyes shut, scrunching their noses, and puckering their lips like a duck. They clamp both hands over their ears, sharply turn their heads to the side, and stomp one foot. Their cheeks are puffed out and their eyebrows are deeply furrowed. Capture a comic moment of stubborn cuteness and mischief, as if they absolutely refuse to listen. Adapt the pose naturally to each subject’s body structure.

Use a soft gray pastel studio, subtle pink rim lighting, smooth and cozy 3D-figure textures, delicate embroidered fabric, and natural skin highlights. Use an 85 mm lens and full-body close framing from a slightly low viewpoint, not straight on. Make it look like a premium 3D animation still, in ultra-high quality.

--ar 9:16 --v 7 --stylize 250 --no text, watermark, extra characters, duplicate subjects, distorted anatomy

What you'll create

Create a premium 3D animation-style still of every subject in a supplied reference image as a tiny, stubborn caricature figure. The figures cover their ears, turn their heads, and stomp in a comic refusal-to-listen pose, set against a soft gray studio with pink rim lighting.

Generation settings

Midjourney v7

Aspect ratio: 9:16
Advanced settings
Negative prompt
text, watermark, extra characters, duplicate subjects, distorted anatomy
Stylize
250

Usage instructions

Provide a reference image containing the subjects you want transformed; the prompt uses it to guide their features and preserve the original subject count. If you want to adjust the look, revise the pink embroidered pinafore details while keeping the instruction to prioritize each subject’s distinctive original clothing.

Frequently asked questions

What should the reference image determine?

Use it to preserve every subject’s distinctive face, hair, skin tone, age impression, and other identifying features, as well as the original subject count. For animals and dolls, preserve their species or defining construction and materials.

Can the outfit differ from the pink embroidered dress described here?

The prompt gives priority to each subject’s distinctive original clothing, so the pink embroidered pinafore-style dress should not replace those identifying outfit details.

Comments

Tried this prompt? Share what you made, compare results, or ask a question

Up to 4 results. Images up to 2 MB; MP4, WebM, or MOV videos up to 4 MB each