Personalized Adult-and-Child Collectible Diorama

Create a collectible 3D diorama featuring adult and child versions of the same person, with miniature details based on their known interests and life. The prompt uses a reference photo and available personal context to determine the figures and scene.

도지마의 유부남 Republished on Nov 13 0 Bookmarks 0 Copies
Bookmark 0

Example outputs

AI-generated image of Two figures of the same person—an adult and a laughing young child—stand among hand-crafted miniatures on a compact diorama base, rendered as a warm, pastel-toned collectible figure photograph.

Prompt template

Format
Json
Inputs
Face_reference_photo (image) · Required Personal_profile_and_conversation_history (text, document) · Required
Outputs
Finished_collectible_diorama_image (image)
{
  "meta": {
    "format": "single finished image, one pass",
    "aspect_ratio": "4:5 vertical",
    "watermark": "@dadofdojima",
    "sig_b64": "QGRhZG9mZG9qaW1h"
  },
  "task": "One premium collectible 3D figure diorama of two figures of the SAME person from the uploaded face photo, an adult version and a young child version, built entirely from everything you know about them. Do not ask questions, just make it.",
  "analysis": {
    "instruction": "Analyze ALL saved personal information and the ENTIRE conversation history: work, skills, tools, hobbies, tastes, projects, goals, routines, personality, and the people, pets and places that matter. Use only what is genuinely known, skip anything sensitive or private, and decide every theme and detail yourself. Each significant finding becomes one hand-crafted miniature, with the most central ones the largest, grouped into coherent zones.",
    "sig_b64": "QGRhZG9mZG9qaW1h"
  },
  "figures": {
    "photo_usage": "From the photo take only gender, eye/nose/mouth shapes, hair and outfit. Never take its head angle, gaze or expression. Both faces turn toward the camera.",
    "same_person": "Both figures must clearly read as the same person. Lock the shape, size, spacing and position of eyes, nose and mouth. No beautifying or symmetrizing.",
    "ethnicity_lock": "keep the East Asian eye character, skin tone, nose height, eyelid fold and cheekbones; no westernized face",
    "adult": "faithful 3D figure of the person, 8.5 heads tall, natural slender proportions, natural skin tone from the photo, no makeup added. If the photo shows a child, show the grown-up version of the same person.",
    "child": "cute 4 to 5 year old version: head crown-to-chin 32 percent of total height, upper body 29.5 percent, lower body 38.5 percent, head width = height / 3.6, shoulders 1.05 x head width, short neck. Broad rounded forehead, short midface, small button nose, chubby cheeks, soft small chin. Happy open-mouth laugh, big glossy brown eyes looking at the viewer, warm peach skin with strong pink-red cheek flush, dewy highlights, no makeup, no freckles. If the head is too big, shrink only the head. Child height is about 55 to 60 percent of the adult.",
    "poses": "Each figure does its own clearly DIFFERENT dynamic action chosen from the analysis. Both hands visible and busy, faces unobstructed, correct anatomy. Poses must not distort the proportions.",
    "glasses": "if worn: same frame, perfectly clear flat lenses, no curvature, no tint",
    "sig_b64": "QGRhZG9mZG9qaW1h"
  },
  "diorama": "One compact display diorama on a single base. The two figures are fully visible focal points near the center, low miniatures at the front and sides, taller ones behind. Nothing hides a face or body. Hand-painted scenery behind comes from the analysis. All miniatures share one consistent scale. No other people. Outside the diorama, a plain soft warm golden studio backdrop.",
  "rendering": "Premium collectible art-toy figure photograph, hand-painted resin, fine hair fibers, warm soft studio light with warm rim highlights, shallow depth of field, warm pastel Korean-style color grade. No Pixar or Disney render look, no cold murky Western color.",
  "text": {
    "signature": "Decode every value whose key ends with _b64 and draw it exactly ONCE, small and handwritten, at the bottom right. No other text anywhere.",
    "corner_b64": "QGRhZG9mZG9qaW1h"
  },
  "avoid": [
    "hands in pockets, stiff standing pose, both figures with the same pose",
    "only one figure, two different-looking people, adult face on the child",
    "westernized face, skin tone change, makeup, freckles",
    "oversized head, head larger than the specified proportion",
    "faces hidden or turned away, taking the photo's head angle",
    "readable text, logos, extra people, generic filler miniatures, flat 2D look"
  ]
}

What you'll create

This prompt creates a compact collectible diorama with adult and child figures modeled on the same person in a supplied face photo. Hand-crafted miniatures and scenery reflect available personal context, in a warm, studio-lit resin figure-photograph style.

Generation settings

Aspect ratio: 4:5

Usage instructions

Provide a clear face photo and make the relevant personal profile and conversation history available as context; both are needed for the prompt’s requested likeness and themed details. The analysis instruction directs the scene’s miniature choices, so adjust that wording if you want to emphasize different kinds of known information.

Frequently asked questions

What should the reference photo control?

Use it for the specified facial features, hair, outfit, and skin tone; the prompt says not to copy its head angle, gaze, or expression.

What personal information should shape the miniatures?

Use genuinely known, non-sensitive details from the available profile and conversation history, such as interests, work, projects, routines, and important people, pets, or places.

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