3.5-Head Grayscale Beatles Street Poster
Create a grayscale, 3D animated-film-style street poster featuring four short-proportioned figures based on the supplied reference photo or photos. The prompt specifies their outfits, expressions, London street setting, title text, and creator credit
Example outputs
Prompt template
- Format
- Json
- Inputs
- Reference_photo (image) · Required
- Outputs
- Finished_vertical_poster (image)
{
"meta": {
"format": "single finished image, one pass",
"aspect_ratio": "2:3 vertical",
"watermark": "@dadofdojima",
"sig_b64": "QGRhZG9mZG9qaW1h",
"color": "black-and-white only",
"proportion_lock": "Every figure is exactly 3.5 heads tall, whatever the photo shows. See the proportion section.",
"priority": "1. the fixed 3.5-head proportions, 2. gender from the photo, 3. face resemblance. These three outrank the poster, the clothes and the scene."
},
"task": "A black-and-white 1960s London street poster: four figures walk side by side toward the camera. In the sky behind them at the upper left, large bold black text reads \"THE BEATLES\".",
"people_rule": "If one person is uploaded, ALL FOUR figures are that same person (the same face and gender in four different outfits and poses). If several people are uploaded, assign them from left to right. Never invent a stranger's face.",
"proportion": {
"unit": "1 head = crown of the hair to the chin, hair volume INCLUDED",
"total": "exactly 3.5 heads from the top of the hair to the soles",
"parts": "head 1; chin to crotch 1 head; crotch to soles 1.5 heads. The crotch sits at about 57 percent of the height from the top.",
"width": "shoulder width about equal to the head width (hair included), a short neck",
"arms": "shoulder to fingertips about 1.3 heads; hanging hands reach the upper thigh",
"clothes": "Coats, jackets and trousers are cut to this short body. Long overcoats end at the knees, and clothes never stretch the torso or legs.",
"photo": "Body proportions are NOT taken from the photo. Only the face comes from the photo.",
"sig_b64": "QGRhZG9mZG9qaW1h"
},
"identity_lock": {
"gender": "Keep the photo person's gender exactly. A female stays clearly female with a feminine face even with mop-top hair and a suit.",
"resemblance": "From the photo take the eyes, nose and mouth, keeping their exact shape, size, spacing and relative position. The figure must be instantly recognizable as the photo person. No beautifying or averaging.",
"ethnicity_lock": "Keep the ethnicity, skin tone value, nose height, eyelid shape and cheekbones. Do not westernize the face.",
"not_from_photo": "head angle, gaze, expression, hair, clothes, body proportions, background"
},
"hair": "all four have the same 1960s mop-top haircut: thick, rounded, with full straight bangs down to the brows, covering the tops of the ears",
"glasses_rule": "Only if glasses are worn in the photo: keep the same frame. The lenses are perfectly flat and clear, with no curvature, no magnification, no glare. Otherwise add none.",
"characters": {
"1_left": "dark suit, white shirt, thin tie, a leather coat over the right arm, looking to the left",
"2": "long dark overcoat over a suit, black tie, arms down, looking at the camera",
"3": "gray tweed jacket over a light sweater, holding a white envelope at the chest",
"4_right": "heavy dark overcoat, black tie, hands in the pockets, looking at the camera"
},
"expressions": {
"rule": "Each figure's expression is drawn from this description, not from the photo. Use the person's own features to make these expressions.",
"1_left": "calm and thoughtful, mouth closed and relaxed, a faintly distant gaze off to the left, no smile",
"2": "deadpan and cool, mouth closed with lips slightly pressed, eyelids a little heavy, a steady unbothered stare at the camera",
"3": "bright and cheerful, a friendly smile with the lips slightly parted and the upper teeth just showing, brows lifted a little, lively eyes looking at the camera",
"4_right": "serious and composed, mouth closed, chin slightly lowered, a calm direct gaze at the camera, no smile"
},
"scene": "Low camera angle, all four shown from head to shoes. A tall brick apartment building on the right, a stone-paved sidewalk, a pale overcast sky. No landmarks.",
"rendering": "A stylized high-end 3D animated-film character render, then shown in grayscale. It is NOT a photograph: no photographic skin, no real-human rendering.",
"final_check": "Before output, measure each figure: total height divided by head height (hair included) must be 3.5, and the crotch must sit at about 57 percent from the top. If a figure is taller than 3.5 heads, shorten the torso and legs; do not enlarge the head. Also check that the gender matches and the face is recognizable as the photo person.",
"text": {
"signature": "Decode every value whose key ends with _b64 and draw it exactly ONCE, small, at the bottom right. No other text besides the title.",
"corner_b64": "QGRhZG9mZG9qaW1h"
},
"avoid": [
"any color, sepia",
"figures taller than 3.5 heads, big head on a normal-length body, long torso, long legs, body stretched by long coats or trousers",
"body proportions copied from the photo",
"gender changed from the photo, male faces for a female photo",
"strangers' faces, faces not resembling the photo",
"photorealistic people, real photograph look",
"head angle or expression copied from the photo",
"Big Ben or other landmarks",
"westernized face, changed skin tone",
"misspelled title, extra text"
]
}
What you'll create
This prompt creates a grayscale 3D animated-film-style poster of four figures walking toward the camera on a London street. It uses reference photo(s) for facial identity and gender while specifying the figures’ proportions, clothing, expressions, and surrounding scene
Generation settings
Usage instructions
Frequently asked questions
How are multiple reference photos assigned to the figures?
When several people are supplied, assign them to the figures from left to right
Which features should the reference photo control?
Use the photo for the person’s face and gender; do not copy its body proportions, hair, clothing, head angle, gaze, expression, or background
How short should the figures be?
Each figure must be exactly 3.5 heads tall, with the crotch at about 57 percent of the height from the top
Comments
Tried this prompt? Share what you made, compare results, or ask a question