Back 中文 CC Prompt Galaxy
Using the provided reference selfie, keep the original photo composition, outfit, lanyard, badge, hand pose, … - AI prompt reference from CC Prompt Galaxy

Selfie face restoration with 7 mini-me Pixar characters AI Image Prompt

Prompt

Using the provided reference selfie, keep the original photo composition, outfit, lanyard, badge, hand pose, indoor ceiling background, perspective, and real-world lighting mostly unchanged. Remove the solid face-obscuring square and reconstruct a plausible natural face for the person, matching the hairstyle, head angle, skin tone, and selfie lighting. Add exactly 7 tiny 3D animated “mini-me” versions of the same person in a cute Pixar-like style, integrated into the scene with realistic contact shadows, depth, reflections, and matching warm light: 1 standing on the left shoulder with arms crossed, 1 standing on top of the badge waving, 1 hugging the right edge of the badge with eyes closed, 1 peeking out from the shirt collar, 1 sitting on the lanyard near the badge in a thinking pose, 1 sitting on the front yellow lanyard strap looking upward, and 1 hanging from the lower lanyard strap with both hands. Make the mini characters wear simplified matching dark clothing and tiny lanyards, interacting naturally with the real objects. Add a subtle magical sparkle atmosphere without changing the realistic selfie base. Overall mood: playful, adorable, and convincing as a miniature world inside the original photo.

AI Image Content Analysis

Content: Structure: subject (person in original selfie), scene (indoor ceiling background), style (realistic selfie + Pixar-like mini characters), composition (keep original), lighting (realistic warm selfie light), color (warm tone), material (real + animated), lens (fixed perspective), proportion (normal person vs mini characters). Key operations: remove solid face square → reconstruct face → add 7 mini characters with specified positions and actions → magical sparkle.

Pros: Positions and actions of mini characters are clearly described for precise generation.

Cons: Heavily relies on reference image; face reconstruction may not be natural.

Reference image: Obvious reference-image dependency

Author
硅基废话
Type
Image
Aspect ratio
3:4
Size
900 × 1200
Added
2026-06-04

Source materials are collected from publicly accessible web pages. Contact us if a rights issue needs review. Copyright & Privacy

Full preview