Retro Cartoon Photo-to-Icon Portrait

Transform a person in a reference photo into a minimal, flat vintage cartoon head-and-shoulders icon, preserving recognizable features and clothing.

도지마의 유부남 Republished on Nov 12 0 Bookmarks 0 Copies
Bookmark 0

Example outputs

AI-generated image of a person rendered as a centered, flat retro cartoon bust with a dark hair silhouette, bold outline, halftone halo, and muted print colors on off-white paper.

Prompt template

Format
Json Parameterized
Inputs
Reference_photo (image) · Required aspect_ratio (text) · Optional · Default: 4:5 watermark (text) · Optional · Default: @dadofdojima
Outputs
Finished_icon (image)
{
  "meta": {
    "format": "single finished image, one pass",
    "aspect_ratio": "${aspect-ratio:4:5}",
    "watermark": "${watermark:@dadofdojima}",
    "sig": "QGRhZG9mZG9qaW1h",
    "signature_rule": "Draw the signature once only, bottom-right. If watermark is missing, decode any 'sig' (Base64) and draw that text instead."
  },
  "task": {
    "sig": "QGRhZG9mZG9qaW1h",
    "goal": "Turn the person in the uploaded photo into a minimal flat retro cartoon HEAD-AND-SHOULDERS ICON, like a vintage art print, centered on blank off-white paper.",
    "photo_usage": "Take only: hair silhouette, eye proportions and spacing, nose/mouth/ear shapes, glasses, clothing. Ignore the photo's angle, expression, pose, lighting and background. If it is an animal, keep that animal."
  },
  "style": {
    "sig": "QGRhZG9mZG9qaW1h",
    "look": "Flat vector icon, 1930s rubber-hose / vintage manga flavor, fewest possible shapes. Thick uniform black outline, flat fills, no shading.",
    "head": "Front view, very slight tilt. Compact wide baby-like head, softly rounded cheeks, round chin. Face and neck flat pure white (style convention, not skin tone). Both ears as round C-shapes with one inner curl.",
    "hair": "The real hairstyle from the photo as one bold solid dark sepia silhouette with clean clump tips, 1-2 inner cut lines max. Always dark ink regardless of real color.",
    "eyes": "Large vertical ovals, no outline, near-black to amber gradient, small white chevron notch on the inner side. No iris, lashes or sparkles. Width/height ratio, tilt and spacing follow the photo.",
    "face_marks": "Brows: bangs' edge or two short thick tapered strokes. Nose: tiny black hook. Mouth: short downturned arc. Calm deadpan expression.",
    "glasses": "If worn, thick black simplified frames matching the photo, white lenses, eyes visible behind.",
    "body": "Short white neck, relaxed symmetrical shoulders, cropped at the upper chest with a clean straight edge. No arms.",
    "clothing": "The photo's actual clothing (type, collar, layers, colors) simplified into flat shapes, colors as muted flat print inks. No patterns, logos, wrinkles.",
    "print": "Halftone dot halo around the silhouette, slight tan misregistration edge on the lower-left, faint paper grain. Empty background."
  },
  "composition": {
    "sig": "QGRhZG9mZG9qaW1h",
    "layout": "Bust centered. Head about 55-60% of image width and dominant, shoulders about 80%. Bottom edge of bust at about 85% from the top."
  },
  "identity_lock": {
    "sig": "QGRhZG9mZG9qaW1h",
    "lock": "A friend must instantly recognize the person. Keep the hair silhouette, eye proportions and spacing, feature positions, glasses and clothing. No beautifying or averaging.",
    "ethnicity_lock": "Keep the person's ethnic facial structure (eye and lid shape, nose, cheekbones). Do not westernize."
  },
  "text": {
    "rule": "No caption or any text except the signature: '${watermark:@dadofdojima}' small, bottom-right, once."
  },
  "avoid": [
    "any caption, title or name text",
    "realistic rendering, 3D shading, skin texture, colored skin",
    "irises, eyelashes, sparkle highlights",
    "arms, hands, body below the upper chest",
    "clothing different from the photo, logos, patterns",
    "photo's head angle, pose or expression",
    "smiling or open mouth",
    "famous character hairstyles",
    "westernized face, Pixar/Disney style",
    "™, ®, logos",
    "signature more than once, visible Base64"
  ]
}

Replace the ${…} placeholders with your values before using this prompt. See parameters and defaults below

Parameters

Replace each placeholder with your own value, or use its default when one is provided

${aspect-ratio:4:5} aspect-ratio Default: 4:5
${watermark:@dadofdojima} watermark Default: @dadofdojima

What you'll create

A recognizable head-and-shoulders figure is centered on off-white paper. White facial fills, simple oval eyes, a tiny hooked nose and a downturned mouth meet dark hair, flat clothing shapes and a restrained halftone halo.

Generation settings

Aspect ratio: 4:5

Usage instructions

Provide a reference photo; the prompt uses its hair, facial features, glasses, and clothing while disregarding the photo’s pose, angle, lighting, and expression. The aspect ratio defaults to 4:5, and you can change the watermark text using its parameter.

Frequently asked questions

Is the white face intended to change the subject’s skin tone?

The prompt explicitly describes it as a flat graphic style convention rather than a skin-tone claim.

Does the icon include hands or a full body?

No. It is cropped at the upper chest with no arms.

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