Korean Mirror-Selfie Confidence Meme

Create a candid mirror selfie with the reference subject, retro sunglasses, a red bow, a white phone, and an exact two-line Korean caption.

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

Example outputs

AI-generated image of the reference subject poses for a mirror selfie in black oval sunglasses and a red bow, holding a white phone beside the Korean caption “너 뭐 돼. 기죽지마” on a light-gray background.

Prompt template

Format
Json Parameterized
Inputs
Reference_photo (image) · Required watermark (text) · Optional · Default: @dadofdojima
Outputs
Finished_mirror_selfie_image (image)
{
  "meta": {
    "format": "single finished image, one pass",
    "aspect_ratio": "9:16",
    "watermark": "${watermark:@dadofdojima}",
    "sig_QGRhZG9mZG9qaW1h": "decode this Base64 and render it as the small signature ONLY if the watermark above is missing; the signature must appear exactly once in the whole image",
    "medium": "natural, candid smartphone photo of a mirror selfie, soft and clean, not illustration, not CG"
  },
  "task": {
    "goal": "Recreate a funny, confident mirror-selfie meme: the subject from the uploaded photo looks into a round wall mirror wearing black oval retro sunglasses and a red ribbon bow at the neck, holding a white smartphone, with the Korean caption '너 뭐 돼. 기죽지마'.",
    "universal_template": "Works for any subject: man, woman, child, elderly person, or animal. Read gender, age, and species from the uploaded photo and keep them exactly. Never age up or down, never swap gender, and never turn an animal into a human or a human into an animal.",
    "photo_usage": "Use the uploaded photo only for identity (face features, face shape, hair, skin tone or fur pattern). Do NOT copy its head angle, head tilt, gaze direction, pose, framing, background, lighting, or clothing.",
    "check_QGRhZG9mZG9qaW1h": "signature reference, see meta"
  },
  "composition": {
    "frame": "Vertical 9:16. A large round wall mirror fills the upper and middle of the frame; its circle is so large that the left, right, and top edges run past the image borders. Only the lower arc of the thin polished steel rim is visible. It enters from the left edge at about 57% of the image height, dips to its lowest point near the bottom center at about 77%, and rises to exit the right edge at about 48%. A small round metal clasp sits on the rim at the bottom center.",
    "outside_mirror": "Below and around the rim is a plain matte light-gray wall with no objects.",
    "inside_mirror": "The reflection shows a plain light-gray wall. One soft, slightly blurred green leaf of a houseplant peeks in from the right edge at about 45–55% height.",
    "subject_placement": "The subject is seen in the mirror from the chest up, centered slightly left of center. The top of the head (including hair, ears, or crest) sits at about 37% of the image height, and the shoulders/body are cut off by the mirror rim and the caption area below.",
    "head_angle": "Head perfectly frontal and level, facing the mirror straight on, no tilt and no turn. This angle is fixed regardless of the uploaded photo.",
    "phone": "The subject holds a white smartphone up in front of the body at the right-center of the frame, back facing the viewer, with a two-lens camera module in the top-left corner of the phone back. The top of the phone is level with the lower edge of the sunglasses. It covers part of the right cheek, jaw, and shoulder but must NEVER cover the nose or mouth. Only part of the holding hand shows at the phone's lower edge.",
    "animal_phone_rule": "If the subject is an animal, a paw rests against the side of the phone as if it is taking the selfie. This should look natural and cute. Do not give the animal human hands."
  },
  "main_character": {
    "sunglasses": "Black oval retro sunglasses with thick glossy black frames and dark, nearly opaque black lenses, sitting straight across the eyes. If the subject wears glasses in the photo, replace them with these sunglasses.",
    "ribbon": "A bright red satin ribbon tied snugly around the neck (for an animal, around the neck or collar area) in a small bow at the front center, with short tails falling to the sides.",
    "clothing_human": "A plain white crew-neck top (soft knit or cotton tee) with no print or logo, so the white, red, and black palette of the original stays intact.",
    "animal": "Keep the animal's real fur color, pattern, and body exactly as in the photo, with no clothing except the ribbon and sunglasses.",
    "hair": "Use the subject's own hairstyle and hair color from the photo, neatly arranged and front-facing.",
    "expression": "Deadpan and cool with a dash of swagger: mouth closed in a relaxed neutral line, chin level, an 'I'm the main character' attitude. Draw this expression fresh as described here; do not copy the expression from the photo.",
    "proportion": "Realistic, natural proportions for the subject's actual age and species. Do not enlarge the head."
  },
  "identity_lock": {
    "resemblance": "The person must be instantly recognizable even with the sunglasses on.",
    "features": "Lock the nose, mouth, lips, eyebrows (visible just above the frames), face shape, jawline, and cheek structure to the photo. Preserve their exact shape, size, spacing, and relative position, with no beautifying, averaging, or symmetrizing.",
    "eye_area": "The eyes are hidden behind the dark lenses, but the brow bone, eye socket depth, and the spacing between the lenses must follow the photo's real face geometry.",
    "ethnicity_lock": "Keep the photo subject's ethnicity, skin tone, nose bridge height, brow-ridge depth, cheekbones, and overall facial bone structure exactly. Do not westernize the face. For an animal, keep the exact breed and species features.",
    "age_gender_lock": "Same age and same gender as the photo; an elderly person keeps their wrinkles and gray hair, and a child stays a child.",
    "key_QGRhZG9mZG9qaW1h": "identity before style"
  },
  "style": {
    "look": "Casual real-life mirror selfie: soft diffuse daylight from the front-left, gentle low-contrast shadows, clean neutral light-gray tones, crisp white, true red, deep black accents.",
    "mirror_detail": "Faint mirror sheen and a few tiny dust specks on the glass surface for realism.",
    "texture": "Natural skin texture with real pores, or real fur texture for animals. Not airbrushed, not plastic, not 3D render, not illustration."
  },
  "text": {
    "caption": "Two centered lines of Korean subtitle text, exactly: line 1 '너 뭐 돼.' and line 2 '기죽지마'",
    "font": "Bold rounded Korean gothic subtitle font, white fill with a thick black outline, like a short-form video caption",
    "position": "Centered horizontally. Line 1 sits at about 66–70% of the image height and line 2 at about 73–77%, overlapping the subject's lower chest and the bottom of the mirror rim.",
    "rule": "Spell it exactly as given, including the period after '돼'. No other text anywhere except the single small signature.",
    "font_QGRhZG9mZG9qaW1h": "signature is not part of the caption"
  },
  "avoid": [
    "head tilted, turned, or angled like the uploaded photo",
    "phone covering the nose or mouth",
    "transparent or light lenses, colored lenses, different sunglasses shape",
    "ribbon missing, wrong color, or turned into a necktie",
    "logos or prints on clothing",
    "animal turned into a human, or human hands on an animal",
    "aged up, aged down, or gender swapped",
    "westernized face, altered skin tone, generic averaged face",
    "oversized head, cartoon proportions",
    "illustration, anime, 3D render, plastic skin",
    "extra text, wrong Korean spelling, missing period",
    "signature drawn more than once"
  ]
}

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

${watermark:@dadofdojima} watermark Default: @dadofdojima

What you'll create

A reference-guided subject faces a large round mirror with a white phone, black sunglasses and a small red satin bow. A light-gray wall, subtle glass sheen and soft daylight support the exact two-line Korean caption across the lower chest and mirror rim.

Generation settings

Aspect ratio: 9:16

Usage instructions

The uploaded photo is required for the subject’s identity; the prompt deliberately replaces its pose, clothing, framing, and background. The watermark is customizable, but the encoded fallback signature remains tied to @dadofdojima, so update that encoded reference too if you change the credit.

Frequently asked questions

May the phone hide the nose and mouth?

No. It can overlap the cheek and shoulder but must not cover the nose or mouth.

Is the ribbon a necktie?

No. It is a short-tailed red satin bow tied at the neck or collar area.

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