Soup-Cooking Art-Toy Portrait from a Reference Photo

Create a portrait-format scene of the person in a reference photo cooking soup at a nighttime outdoor station, rendered as a soft 2.5D/3D designer art-toy figure. The prompt specifies the cook’s likeness, pose, outfit, expression, and a detailed campfire cooking setup.

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

Example outputs

AI-generated image of a reference-based art-toy figure stirs creamy yellow soup in a blue enamel pot beside a stone fire pit at dusk, with a lantern, warm amber light, and a softly blurred grassy background.

Prompt template

Format
Json
Inputs
Reference_photo (image) · Required
Outputs
Finished_image (image)
{
  "meta": {
    "format": "single finished image, one pass",
    "aspect_ratio": "3:4 portrait",
    "watermark": "@dadofdojima",
    "sig_b64": "QGRhZG9mZG9qaW1h",
    "signature_rule": "Decode every *_b64 value (same handle). Draw it exactly ONCE, small, handwritten white, bottom-right corner. No other signature."
  },
  "face_composite": {
    "priority": "FACE COMPOSITE: the person in the uploaded photo IS the soup cook. Looking exactly like that person is the top priority.",
    "take": "Eye size and shape, eyelids, eyebrows, nose, lips, face roundness, cheek fullness, skin tone, hair (length, style, color, bangs), apparent gender. No beautifying or generic face.",
    "do_not_take": "ONLY expression, gaze, head angle, accessories, clothes, background.",
    "ethnicity_lock": "Keep the person's ethnicity and facial structure. Never westernize.",
    "source_b64": "QGRhZG9mZG9qaW1h"
  },
  "scene": {
    "goal": "Close, large view of ONE soup station at night, real photographic look. Only this station and the one cook. Nothing else.",
    "camera": "Elevated 30 degrees, pot, fire pit and cook fill 75% of the frame, face large enough to read both eyes. Plain grass softly out of focus elsewhere. Dusk.",
    "station": "Grey stone fire pit with a small wood fire, large deep-blue enamel pot of smooth creamy yellow soup with steam, tall round blue gauge with a red wedge behind it, wooden sign reading 115 at the left, low wooden step at the right with a glowing lantern on a stool.",
    "light": "Warm amber lantern and fire glow, soft blue-violet dusk, warm clean color.",
    "scene_b64": "QGRhZG9mZG9qaW1h"
  },
  "cook": {
    "pose": "Stands on the step behind-right of the pot, stirring with a long wooden ladle in both hands, leaning slightly toward the pot.",
    "outfit": "Wide straw hat with a small pale-blue wing ornament, brim tipped back so it never shades the eyes; plain cream linen shirt, dark brown trousers. No accessories, no makeup.",
    "proportion": "3.3 heads tall with hair. Head about 30% of height, wider than shoulders, chin to hip about 1 head, shoulder to wrist about 0.9 head, hip to sole about 1.3 heads, short soft legs, small hands and feet.",
    "age_rule": "Follow the photo person's age group. Adult: young-child-looking art-toy figure with the proportion above. Child: no transformation, natural proportions for that age.",
    "render": "Korean-style soft 2.5D/3D CG designer art-toy figure, smooth matte skin, clean 4K detail, warm pastel color. Lit by the same lantern light. Figure looks small next to real-size props."
  },
  "expression_gaze": {
    "expression": "From this text only, never the photo: calm, focused, mouth closed in a neutral line, relaxed brows and cheeks, no smile.",
    "gaze": "Head turned only about 20 degrees toward the pot, both eyes fully visible, eyes lowered toward the soup. Never looks at the camera.",
    "lock_b64": "QGRhZG9mZG9qaW1h"
  },
  "text": "Only the text 115 on the sign.",
  "avoid": [
    "a different or generic face, different gender or hairstyle than the photo",
    "copying the photo's expression, smile, head angle or gaze",
    "eye contact with the camera, a full side profile, shadowed eyes",
    "accessories, makeup, earrings",
    "taller than 3.3 heads, long limbs",
    "cabbage, corn, carrots, other people, animals, or any prop not listed",
    "westernized face, changed skin tone, cold muddy color",
    "any second signature or watermark"
  ],
  "final_check": "Compare the face with the photo and redraw if it is not the same person. Check both eyes are visible, the expression is neutral, the gaze is on the soup, and the figure is 3.3 heads."
}

What you'll create

Create a reference-based art-toy figure stirring soup at an outdoor cooking station at dusk. The scene combines a soft 2.5D/3D CG figure with real-size props, warm lantern and fire glow, and a softly out-of-focus grass background.

Generation settings

Aspect ratio: 3:4

Usage instructions

Provide a clear reference photo; the cook’s facial identity is based on that image, while the expression and gaze follow the prompt’s written directions. To change the setting or figure styling, edit the corresponding `scene` or `cook` fields while preserving the specified face, gaze, and text constraints.

Frequently asked questions

What does the reference photo control?

Use it to match the person’s facial features, skin tone, hair, and apparent gender. Do not take the photo’s expression, gaze, head angle, accessories, clothes, or background.

How does the figure’s age affect its proportions?

Follow the age group in the photo: adults become young-child-looking art-toy figures with the specified 3.3-head proportion, while children keep natural proportions for their age.

Which text should appear in the scene?

The sign should read “115.” The prompt also specifies a small, handwritten white “@dadofdojima” watermark in the bottom-right corner.

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