GPT Image 2Prompts

Please convert the input image into a minimalist single-line illustration suitable for social media avatars, with a default aspect ratio of 1:1. The subject can be a person, cat, dog, other animals, or plants. Based on the input subject type, automatically extract the most recognizable outer contours and key features, and summarize them into an artistic, minimalist single-line avatar/illustration. The overall goal is not mechanical tracing, but retaining the subject's spirit and recognizability, reconstructing it with a single line in a more advanced and spirited way. [Subject Processing Logic] If the subject is a person: Retain key contours like face shape, forehead, nose bridge, mouth turns, chin, ears, and hairstyle. Keep only a few necessary facial features, avoiding fragmented details. If the subject is a cat, dog, or other animal: Retain key recognizable features like head shape, ears, muzzle contour, eye expression, and the general flow of fur, avoiding complex fur details. If the subject is a plant: Retain the most recognizable morphological features of flowers, leaves, branches, fruits, or overall contours, using simple lines to summarize its posture and rhythm. [Line Style Requirements] Lines must be smooth, rounded, stretched, expressive, and natural, as continuous as possible, done in one breath, with a sense of breathing and improvisation. The overall style should be dominated by soft curves, rarely using sharp, stiff, or rigid broken lines. Do not draw the contour as a rigid logo-style stroke, do not be mechanical, stiff, fragmented, or overly neat. Lines should be simple but full of vitality, with a bit of casualness and artistic temperament, looking relaxed, agile, and collectible. [Composition and Avatar Adaptation Logic] The overall composition is designed as a 1:1 square. The subject should be clear, concentrated, and striking, suitable for use as a social media avatar. It is recommended to center or slightly off-center the subject, with complete contours and a clear visual center of gravity. Please leave reasonable blank space to avoid the subject being too large and touching the edges, ensuring that even after a circular avatar crop, the key parts of the subject remain complete, clear, and highly recognizable. The image should remain simple, powerful, and easy to recognize at small sizes. [Background Color Logic] The background color should not be fixed to a single color, nor default to blue. Please prioritize extracting a "dark tone logic" suitable for the background based on the comprehensive color temperament of the input image itself, selecting an appropriate background color from the dark tendencies related to the subject. If the user explicitly specifies a background color, prioritize using the user-specified color, and on this basis, maintain an overall advanced, restrained, and tense dark tone treatment. Background color selection should follow these principles: - Mainly use dark, pure, solid color backgrounds with emotional tension. - The background should effectively set off the subject's lines, forming a clear contrast. - Do not fix a certain hue; flexibly choose dark tones like dark blue, dark green, dark purple, dark brown, dark gray, burgundy, ink black, blue-gray, dark olive, etc., based on the image's temperament. - If the image itself suits a cool background, use a cooler dark logic. - If the image itself suits a warm background, use a warmer dark logic. - If the user specifies a color, choose a more composed, darker version of that color suitable for avatar display. - Avoid fluorescent, cheap, bright feelings. - Avoid dull, dirty, low-contrast, spiritless backgrounds. [Subject Line Color Logic] The main contour line of the subject is recommended to be a thin white line, or a thin light-colored line that forms a sharp contrast with the background. The line should not be too thick, nor so thin that it loses its presence; it should be light, clear, and clean. If the background is dark, prioritize using white or light warm white lines to enhance recognizability and freshness. [Decorative Element Logic] Add a small amount of random, improvised, natural small colored decorative elements around the subject to enhance agility, fun, and relaxation. Decorative elements can be: small dots, small arcs, small stars, small smiley faces, short strokes, small lightning bolts, simple geometric symbols, etc. The decorative style should be rounded, playful, relaxed, and natural, like casually added inspirational decorations, rather than deliberately laid out template elements. Decorations must be well-controlled: - Small in quantity, just enough. - Naturally distributed, slightly random. - Not too dense, not too rigidly symmetrical. - Do not overshadow the main subject. - Exist only as an auxiliary atmosphere. - Decorative colors can be extracted in small amounts from the adjacent or complementary color logic of the background color, or use a small amount of pop colors, such as yellow, orange, pink, light purple, light coral, etc. - The overall decoration should coordinate with the subject and background, not looking like a stiff collage. [Overall Style Goal] The overall picture should be minimalist, light, modern, agile, and artistic. It should have both the clean feel of a single-line illustration and the relaxation and vitality of an improvised doodle. Make the subject look rounded, stretched, and spirited, rather than stiff, dull, or templated. The final effect should be suitable as a social media avatar: striking, recognizable, interesting, well-designed, while possessing a relaxed, happy, naturally vivid, and collectible artistic temperament. [Problems to Avoid] No realism, no shadows, no light and dark volumes, no complex textures, no complex scenes. No mechanical tracing, no stiffness, no sharpness, no over-neatness. Do not fix the background color to a single blue template. Do not make the background too bright and glaring, nor dull and spiritless. Do not over-decorate, do not steal the spotlight from the subject. Do not lose agility due to minimalism, nor lose restraint due to decoration. Ratio 16:9

Create a vertical close-medium authentic daily-life photograph of {argument name="subject" default="a young Japanese woman in her early 20s"} sitting sideways on the edge of an unmade bed in a small bedroom on an ordinary morning. Shoot from a slightly elevated three-quarter angle, as if captured candidly, with her posture relaxed and sleepy, leaning forward a little, knees together, one hand resting on her knee and the other loosely holding a dark smartphone near her thigh. Her face is intentionally obscured by a soft rectangular privacy blur, while her naturally straight black hair is loosely gathered and messy from sleep, with stray strands around her cheek and neck. She wears {argument name="sleepwear" default="an oversized loose white cotton T-shirt and light gray soft shorts"}, unstyled and casual. The room feels lived-in: there is exactly 1 large window on the left with bright morning daylight, exactly 1 potted plant on the windowsill, exactly 1 small green bottle near the window, exactly 1 cluttered bedside table with small bottles, jars, and daily items, and an unmade bed with cream pillows and {argument name="blanket color" default="muted sage green"} rumpled bedding. Use soft directional morning window light from the left, a cool-to-warm transition, visible dust and haze, gentle shadows across the shirt and legs, realistic skin texture on the hands and collarbone, natural fabric wrinkles, shallow depth of field, muted colors, slight grain, documentary realism, no posing, no glamour styling, no makeup, no artificial studio lighting.

Create a vertical smartphone-style food photo of {argument name="dish" default="homemade takoyaki"} served on a plain round white ceramic plate on a gray stainless-steel kitchen counter. The takoyaki is piled high and looks glossy and appetizing, covered with dark brown takoyaki sauce, thick pale yellow mayonnaise zigzags, fluttery bonito flakes, green aonori seasoning, and sliced green onion rings spilling over the plate edge. Add cute hand-drawn white doodle illustrations over the photo as if drawn with a marker: exactly 2 round octopus mascot characters wearing festival headbands, one large octopus peeking from behind the upper right of the plate holding 1 takoyaki skewer and looking surprised, and one smaller octopus at the lower left leaning against the plate; include exactly 2 small doodled takoyaki balls, one circled near the bottom center of the plate and one smiling ball at the lower right. Surround the dish with playful white outline decorations, including approximately 11 stars, 5 hearts, 3 flower-like puffs, 4 small bubbles or circles, and several motion/splash marks, all floating around the food without covering the main toppings. Keep the lighting casual and realistic, with shallow phone-camera perspective from slightly above, warm indoor color, crisp food texture, and a whimsical cozy social-media aesthetic. No printed text or labels.

Create a {argument name="style" default="clean, minimal, high-end"} facial beauty report based on this photo. Use a black-on-white design with thin lines, rounded cards, and a luxury aesthetic. Include a simple contour line drawing of the face, an honest attractiveness analysis (symmetry, proportions, bone structure, skin, etc.), clear scores, strengths, areas for improvement, and actionable grooming/style recommendations. Keep it data-driven, visually refined, and not overly flattering. Make the aspect ratio {argument name="aspect ratio" default="9:16"}.

Create a cinematic photorealistic urban night scene in a narrow wet brick alley after rain, with exactly two young anime-inspired cat-eared female street dancers as the main subjects, both faces obscured by soft square blur blocks. The left dancer wears a {argument name="left outfit color" default="white"} zip-up tracksuit with black triple side stripes, white sneakers, a small black paw-print emblem on the chest, black cat ears, and very long black hair with red streaks, red bead tassels, and floral hair ornaments; she is squatting low in a dynamic breakdance pose with one leg kicked toward the camera, one arm raised in a hand sign, and the other arm bent across her face. The right dancer wears a {argument name="right outfit color" default="black"} zip-up tracksuit with white triple side stripes, black chunky shoes, a white paw-print emblem on the chest, white cat ears, and extremely long flowing silver-white hair with black floral hair ornaments; she mirrors the same low squat dance pose with one leg extended forward, one arm raised in a hand sign, and the other arm bent across her face. Include exactly five blurred background onlookers in dark streetwear: one standing in the center distance, one crouching behind the dancers, one small figure between them, and two figures on the far right. Place a dark old sedan on the left with bright headlights and a visible Russian-style license plate, beside graffiti reading {argument name="graffiti text" default="HARD BASS"} and {argument name="secondary graffiti" default="CYKA"} with a simple cat doodle. The alley should have black brick walls, metal fire escapes, overhead industrial lamps, misty backlight, puddles, glossy reflections, dramatic rim lighting, shallow depth of field, motion blur in the kicking shoes and flying hair, high-contrast cyberpunk street photography, ultra-detailed fabric folds, wet asphalt texture, and a low wide-angle camera viewpoint at ground level. Use a gritty but polished 2026 AI-image aesthetic, realistic lighting, 4:3 composition, no extra main characters, no readable text besides the graffiti and license plate.

Create a vertical vintage travel poster illustration for {argument name="destination headline" default="NIHONBASHI"}, with the subtitle {argument name="location subtitle" default="TOKYO, JAPAN"} centered beneath it in thin uppercase lettering, flanked by short horizontal rules. Use an elegant cream paper background and monochrome red ink line art, like a refined architectural pen-and-ink city guide poster with subtle manga-inspired human figures. The scene shows a wide pedestrian street in Nihonbashi, Tokyo, viewed in deep one-point perspective between tall modern and historic commercial buildings. On the right foreground, place the ornate Nihonbashi bridge guardian statue: a highly detailed seated winged kirin/lion-like creature on a stone pedestal beside a decorative streetlamp, with the pedestal plaque reading {argument name="bridge plaque text" default="日本橋"}. A canal or bridge railing runs behind it, with trees lining the water. The foreground includes a crosswalk, short bollards, tactile paving, and pedestrians walking in business and casual clothes; include a man with a bicycle at the right edge, a woman in a long skirt crossing near the statue, a suited man walking away in the center, and a woman holding a shopping bag at the left. The left side has storefronts and awnings in detailed perspective. Include exactly 7 visible text elements: the large top headline “NIHONBASHI”, the subtitle “TOKYO, JAPAN”, a vertical hanging banner reading “NIHONBASHI”, a small sign reading “COFFEE”, an awning reading “BOOKS”, a small awning reading “BAKERY”, and a vertical sign reading “CAFE”, plus the stone bridge plaque text “日本橋” if using the default. Keep all architecture, people, signs, trees, railings, lamps, and paving rendered as delicate red outlines with minimal shading, spacious white negative areas, crisp poster composition, and a calm sophisticated Tokyo travel atmosphere.

Create a vertically stacked comparison image with exactly 2 horizontal sections: the top section is a minimalist studio product shot of 1 small enamel refrigerator magnet, centered on a pale textured blue background with lots of negative space; the magnet is shaped like a miniature mountain landscape souvenir, with a shiny gold-metal outline, raised enamel details, snow-capped blue-gray peaks, green rolling hills, dark evergreen trees, and a foreground band of tiny yellow, purple, and blue wildflowers, casting a soft realistic shadow. The bottom section is a bright realistic landscape photo of {argument name="location" default="an alpine grassland valley"} under a blue sky with fluffy white clouds: wide green rolling hills, distant snow-streaked mountains, dark pine forests, and a colorful wildflower meadow in the foreground with yellow, purple, and blue blooms. Include exactly 4 horseback riders crossing the meadow near the lower center, small in scale against the vast scenery. Make the magnet in the top section feel like a distilled souvenir version of the landscape below, preserving the same key visual elements: mountains, green hills, conifer trees, wildflowers, and open sky. Use clean editorial composition, realistic lighting, crisp detail, soft shadows, and a calm travel-memory aesthetic. No visible text, no labels, no watermark.

Create a high-end studio product photograph of a glossy anime prize figure of {argument name="character name" default="Reze"}, shown as a seated young woman figurine kneeling on a round tufted cushion base. The figure has {argument name="hair color" default="dark purple-black"} short bobbed hair with layered bangs and a small low side bun, pale skin, and a deliberately soft blurred face with no readable facial details. She wears a sleeveless pale lavender-white button-up blouse with a small black ribbon tie at the collar, fitted high-waisted black shorts, and smooth black thigh-high stockings; the blouse has faint semi-transparent lettering across the chest reading {argument name="shirt text" default="NOMAI"}. Pose her with both knees folded forward on the cushion, one hand resting on each side for support, torso slightly leaning back, shoulders relaxed, creating the look of a collectible scale figure rather than a real person. The base is a single circular display plinth: a black outer rim with a tan caramel-brown quilted leather-like cushion on top, including deep button-tufted depressions and glossy highlights. Use clean white seamless background, soft diffused studio lighting, sharp focus on the figure and base, realistic PVC figure material, subtle skin and fabric sheen, premium prize-figure product photography, three-quarter front view from slightly above, centered composition, no packaging, no extra props, no watermark except the faint shirt text.

Based on a real photo uploaded by the user, generate a 2x2 grid of "New Chinese Qipao Retro Mother Portraits". The four small images must feature the exact same woman, strictly preserving her face shape, facial proportions, eyes, nose, lips, skin tone, {argument name="sense of age" default="50-65 years old"}, basic hairstyle, body type, and overall temperament. Do not turn her into a stranger, do not make her look younger, do not use internet-celebrity filters, and do not over-smooth the skin. The overall style is a retro studio portrait in a New Chinese Qipao. The image should be exquisite and natural, with the feel of a traditional photo studio, retro Chinese magazine photography, a slight film grain, and a soft oil painting texture. The clothing should be an elegant and decent modified Qipao or New Chinese suit, using colors like {argument name="clothing color" default="dark green, navy blue, burgundy, pearl white, lotus root pink"}. Props can include folding fans, round fans, white magnolias, orchids, pearl necklaces, wooden chairs, wooden tables, and small vases. The background can be uniformly {argument name="background color" default="dark green, black, misty blue, dark purple, retro red, or warm caramel"}. The four small images must have distinct variations: Top left: Full-body or three-quarter seated posture, the mother sitting on a retro wooden chair, holding a folding fan, looking at the camera with a gentle smile. Top right: Close-up half-body, the mother wearing a Qipao, holding flowers, body turned slightly sideways to the camera, looking off-camera with a calm expression. Bottom left: Side profile composition, the mother looking down with a slight smile or looking at a flower branch, gently tucking hair behind her ear with one hand, creating a gentle and quiet scene. Bottom right: Frontal half-body shot, the mother sitting elegantly, hands naturally folded, with a wooden chair, vase, and soft retro lighting in the background, showing a warm and natural expression. The actions, expressions, gestures, head turns, framing, and shooting angles of the four images must be different. The background, clothing color, clothing style, flowers, props, jewelry, hairstyle, set, and main lighting style must remain consistent. The lighting should be soft, the skin real and natural, retaining the mother's mature temperament and slight traces of age. Avoid cheap photo studio styles, heavy makeup, and excessive gorgeousness.

Generate an illustration of "me" as you imagine it. Features include a {argument name="style" default="Japanese illustration style"}, distinct character features, natural emotional expressions, a half-body composition, dynamic poses, exquisite clothing details, a {argument name="texture" default="hand-drawn graffiti style"}, ink splatter strokes, free-flowing lines, a blend of pastels and ink, a comic sketch texture, a {argument name="background" default="minimalist white background"}, surrounding symbolic elements, a strong atmosphere, high detail, and high quality.

Using REFERENCE_0 as the character base, transform the angel character sheet into a cute chibi three-view turnaround. Keep the original angel design, white feather wings, long pale hair, flower hair accessory, white dress, blue ribbons, barefoot look, soft pastel anime rendering, and plain white background. Change the body proportions to super-deformed chibi style with a large head, small childlike body, simplified hands and feet, and rounder silhouette while preserving the same delicate details. Arrange exactly 3 full-body views horizontally: front view on the left, side profile in the center, and back view on the right. Maintain clean character-reference-sheet spacing, high detail, soft lighting, and 4K-quality finish.

Write an exchange diary entry. I want you to create an image with a message directed to {argument name="Target" default="Current me"}. It should look like an exchange notebook used by elementary school students, written in pencil on a {argument name="Size" default="B5"} size notebook. Make the handwriting irregular, like an elementary school student's handwriting.

Create an authentic candid daily-life photograph at a lively night market at night, showing a three-quarter body shot of two young women interacting mid-gesture beside glowing food stalls. The woman on the left is an {argument name="left woman ethnicity" default="East Asian"} young woman in her early 20s with messy dark hair tied in a loose bun, wearing an oversized light gray zip-up hoodie with drawstrings and a black shoulder bag strap; her face is anonymized by a flat opaque dark-brown rectangular block. She stands slightly turned toward the right, holding out one small bamboo skewer with a bite-sized piece of grilled street food at the tip, arm extended at chest height in a casual offering gesture. The woman on the right is an {argument name="right woman ethnicity" default="African"} young woman in her early 20s with a high natural curly puff, wearing a black bomber jacket over a white basic T-shirt and carrying a black shoulder bag; her face is also anonymized by a flat opaque dark-brown rectangular block. She leans slightly toward the skewer, one hand raised with fingers loosely open as if about to accept the food, body language suggesting surprise, amusement, and appetite. Use warm amber stall lighting from the left, visible specular highlights on clothing and skin, deep night shadows, and a strong cinematic contrast between orange food-stall glow and cool blue background lights. The background should include one busy street-food stall on the left with trays of food, metal framing, bright exposed bulbs, and blurred shoppers; on the right and far background include dark market silhouettes, neon signage, bokeh lights, and shallow depth of field. Compose from a three-quarter side angle, waist-up to thigh-up framing, natural handheld street photography, realistic lens perspective, candid timing, slight motion in stray hairs and hands, high detail, photorealistic texture, documentary style, no posing, no text.

Using the provided reference photo as the source plant, transform it into a clean plant-identification infographic poster. Isolate the fuzzy green spherical Dianthus stems from the busy street/florist background, retouch them into a crisp studio botanical cutout, and place them on a soft pale green botanical layout with rounded information cards, leaf-line decorations, small green icons, and a trustworthy plant-app aesthetic. Use Simplified Chinese typography. Main headline: {argument name="headline text" default="绿毛球石竹"}; Latin name subtitle: {argument name="latin name" default="Dianthus barbatus ‘Green Trick’"}; show a confidence badge at the upper right reading high-confidence identification. Add exactly 9 content panels: 1 top-left identification recommendation summary, 2 classification judgment, 3 key identification evidence, 4 similar plant distinction, 5 basic file, 6 growth habits and care, 7 risk warning, 8 how to confirm, and 9 bottom fun-fact strip. Keep the infographic text dense but neatly aligned, with numbered green circle markers for panels 2 through 8, icons matching each topic, and the plant image as the central visual focus. Plant identity to present: {argument name="identified plant" default="Dianthus barbatus ‘Green Trick’, also called Green Trick"}; family/genus line: {argument name="taxonomy line" default="Caryophyllaceae | Dianthus | sweet william / Dianthus barbatus horticultural cultivar"}.

Ultra-realistic 16:9 close-up studio photograph of a {argument name="country" default="Russian Federation"} citizenship identity card placed on a clean light grey or white background. The card is horizontally aligned, centered in frame, captured in sharp macro focus with soft diffused studio lighting. The design follows an official government-issued ID aesthetic with structured typography, layered anti-counterfeit detailing, and engraved security graphics inspired by modern national identity cards. Large bilingual header at top reads: “РОССИЙСКАЯ ФЕДЕРАЦИЯ” with secondary title “КАРТА ГРАЖДАНИНА / CITIZENSHIP CARD.” Top-left corner features the {argument name="coat of arms" default="Russian double-headed eagle"} coat of arms on a red shield background. Upper-right includes the national flag. The card contains a passport-style portrait of a {argument name="identity" default="young Russian woman"}: – Fair Eastern European skin tone with subtle natural freckles – Medium-length ash brown to dark blonde hair with soft waves – Grey-blue eyes, neutral expression, direct gaze – Minimal makeup, realistic facial texture – Wearing a dark navy blazer or black top with subtle gold jewelry Personal identity information displayed in Russian and English: – Фамилия / Surname: ИВАНОВА / IVANOVA – Имя / Given Names: АНАСТАСИЯ СЕРГЕЕВНА / ANASTASIA – Дата рождения / Date of Birth: 12.03.1992 – Место рождения / Place of Birth: МОСКВА / MOSCOW – Гражданство / Nationality: РОССИЙСКАЯ ФЕДЕРАЦИЯ / RUSSIAN FEDERATION – № гражданства / Citizenship No.: RUS-1234567 – Дата выдачи / Date of Issue: 20.05.2024 – Действителен до / Date of Expiry: 20.05.2034 – Пол / Sex: Ж / F Security and design features include: – Intricate guilloche security patterns across the entire card surface – Fine microprinted anti-counterfeit line textures – Holographic foil patches with iridescent rainbow reflections – Transparent layered security overlays – Circular “RUS” holographic security seal – Large engraved illustration of the Kremlin towers and Saint Basil’s Cathedral across the right side – Faint Russian double-headed eagle watermark integrated into background – Machine-readable identity strip at the bottom edge – Signature at bottom: “Иванова А. С.” Additional design styling: – Soft red, blue, and silver-grey gradients inspired by the Russian flag – Polycarbonate laminated material with realistic reflections and embossed texture – High-security official document aesthetic with engraved banknote-style artwork Lighting is soft and evenly diffused, emphasizing holographic effects, engraved textures, and realistic print detail without harsh shadows. No hand visible. Style: ultra-detailed, photorealistic, 8K resolution, macro clarity, professional product photography, documentary realism, highly realistic government ID card design.

Create a bright anime-style YouTube thumbnail illustration in 16:9 landscape format, set at the entrance of a Japanese school during spring cherry blossom season. Show exactly 4 high-school girls standing side by side on the right half of the image, each wearing the same uniform: dark green sailor-style blazer with white trim, large burgundy bow, mustard-yellow pleated skirt, white socks, brown loafers, and each holding a brown leather school satchel in front of her. Give the 4 girls distinct hairstyles and hair colors: long flowing pink hair, short wavy light brown hair, long blonde hair, and long straight black hair. Keep their faces softly blurred or featureless so the thumbnail emphasizes the concept rather than individual identities. The background should feature a school gate, school building hints, blue sky, sunlight, abundant pink cherry blossom trees, and many falling sakura petals, with a polished colorful anime-game look, high saturation, crisp lighting, and cheerful romantic spring atmosphere. On the left half, add large bold Japanese thumbnail text reading {argument name="headline text" default="もっと！モット！ときめき'99"}, using thick rounded lettering, white fill with vivid pink outline and warm orange shadow, arranged dynamically with oversized emphasis on “モット!” and “'99”. Add small decorative hearts, sparkles, and an underline swoosh around the text. Compose it like a clickable nostalgic dating-sim themed thumbnail: characters on the right, text dominating the left, energetic diagonal layout, glossy highlights, clean readable typography, cinematic depth, floating petals in the foreground, no extra characters, no logos, no watermark.

{"type":"luxury 3D pop-up storybook illustration","scene":"An open hardcover fairy-tale picture book on an antique dark wooden desk, photographed from a slightly high front angle, transformed into an elaborate layered paper-theater diorama. Warm candlelit ambiance, shallow depth of field, ornate vintage fantasy mood, soft golden bokeh, roses and petals scattered around, highly detailed paper cutouts with white die-cut edges, embossed gold foil, delicate shadows between layers.","main_subject":{"description":"A single anime-style young princess character popping out from the center of the open book as a life-size paper cutout, seated gracefully in a voluminous sparkling red ball gown with many translucent ruffles, gold trim, off-shoulder bodice, necklace, long wavy brown hair, elegant pose with one hand near her face. Face area should be softly obscured or left as a customizable portrait placeholder.","character_name":"{argument name=\"character name\" default=\"Nozomu\"}","dress_color":"{argument name=\"dress color\" default=\"deep ruby red\"}","storybook_title":"{argument name=\"storybook title\" default=\"Welcome to Nozomu's World\"}","subtitle":"{argument name=\"subtitle\" default=\"The Beginning of Dreams and Magic\"}"},"layout":{"format":"single full-page vertical image, realistic handcrafted pop-up book sculpture","discrete_labeled_text_areas_count":6,"labeled_text_areas":[{"position":"upper left arched banner","text":"{argument name=\"storybook title\" default=\"Welcome to Nozomu's World\"}","secondary_text":"The Beginning of Dreams and Magic","style":"cream parchment sign with scalloped edge, roses, gold filigree"},{"position":"left middle plaque","text":"Nozomu","secondary_text":"Kind and hardworking. Loves talking to the stars in her dreams.","style":"small ornate cream plaque surrounded by red roses"},{"position":"lower left book page","text":"Once upon a time, there was a girl named Nozomu. Nozomu loved red dresses. At night, she would softly speak to the moon.","style":"printed story text with a large decorative initial character"},{"position":"right middle standing chapter card","text":"Chapter 1: Nozomu and the Night of the Red Dress","secondary_text":"A small promise watched over by the moon.","style":"upright framed card with roses and ribbon"},{"position":"lower right book page","text":"'Someday, I want to do something wonderful that warms someone's heart and makes them smile.' The stars twinkled, watching over that dream.","style":"cream printed page with gold ornament and story text"},{"position":"bottom center ribbon label","text":"Nozomu's Story","style":"curled red ribbon banner across the bottom edge of the open book"}],"pop_up_elements_count":10,"pop_up_elements":["central princess cutout in red gown","large moon-and-stars night-sky arch behind the character","cream-and-gold sofa on the right","tall arched palace window on the far right","purple fairy-tale castle on the left","layered paper staircase/path rising from the book gutter","small round tea table with roses near the sofa","golden pumpkin carriage illustration on lower right page","miniature red dress illustration on lower left page","rose and cloud paper clusters throughout"],"background_props_count":5,"background_props":["lit lantern or candle on the left","antique book or framed cover in the rear right reading Conte de Rêves","pink roses in the dark background","lace doily under the book","loose rose petals on the tabletop"]},"style":{"art_direction":"romantic Japanese shoujo anime blended with premium paper craft, pop-up book engineering, Victorian fairy tale, ornate gold accents, dreamy magical realism","color_palette":"warm ivory parchment, antique gold, ruby red, blush pink roses, navy starry sky, dark walnut wood","lighting":"warm candlelight from left, soft glowing highlights on gold foil, gentle vignette","render_quality":"ultra detailed, crisp paper textures, layered depth, realistic shadows, glossy ink, high-resolution product-photography finish"},"constraints":"No modern objects, no flat poster look, avoid messy unreadable typography except decorative small story text, preserve the sense that every object is made from layered printed paper rising from the open book."}

# Conceptual Crowd Portrait Please generate a highly visually striking "Conceptual Crowd Portrait" based on a clear real-person photo uploaded by the user. ## I. Character Identity Requirements Please strictly refer to the real identity features of the person in the uploaded photo, including: - Facial contour - Facial feature proportions - Eyebrow and eye structure - Nose shape - Lip shape - Basic hairstyle - Skin tone and sense of age - Eye expression and temperament - Overall character recognizability The final image must make it recognizable at a glance that it is the same person as in the uploaded photo. Do not change them into a stranger, do not westernize, do not make them look like an internet celebrity, do not over-beautify, and do not generate a fake AI face. ## II. Core Creativity Recreate the character portrait as a large-scale conceptual installation art piece. The entire portrait head is composed of thousands of miniature people. These tiny figures are densely arranged like a crowd formation, jointly spelling out the face, hair, facial features, glasses, beard, clothing outline, and overall demeanor of the person in the uploaded photo through different standing positions, clothing shades, crowd density, and shadow changes. The picture looks like a huge portrait made of a real crowd, and also like a contemporary art installation placed in a pure white space. ## III. Composition - Vertical composition, suitable as a visual poster or conceptual cover - Aerial view / Isometric top-down view - The character's head occupies the main body of the picture - The background is a minimalist pure white space, retaining a lot of blank space - A small number of tiny people stand scattered around the portrait, enhancing the sense of spatial proportion and real scale - The crowd as a whole forms a clear character outline, and the position of the facial features must be accurate ## IV. Visual Elements ### Crowd Structure - Use dense crowds to form hair, eyebrows, eyes, bridge of the nose, lips, jawline, and clothes - Dark-colored crowds form hair, glasses, collars, and shadow areas - Light-colored crowds form facial highlights and skin areas - The crowd arrangement should be natural, with a real sense of crowding and tiny random changes - Each tiny person must have an independent posture, clothing, and direction, not like copy and paste ### Light and Shadow Effects - Soft cinematic lighting - Low-angle long shadows - Natural shadows cast under the feet of the crowd - Unified shadow direction, enhancing spatial realism - The overall picture is clean, premium, and quiet ### Color Style - Warm neutral tones - Mainly off-white, light brown, gray, and charcoal black - Dark clothing is used to strengthen the character's outline - Skin tone areas transition with light beige and warm gray crowds - Overall color restraint, do not be overly bright ## V. Art Style Overall style reference: - Contemporary conceptual art - Museum-level installation art - Hyper-realistic crowd simulation - Minimalist visual poster - Premium digital art - Social metaphor visual works - Cinematic conceptual photography The picture must have a strong sense of visual metaphor: The image of a person is composed of countless tiny individuals, which has both the shock of a macro portrait and the explorability of micro crowd details. ## VI. Mood Overall atmosphere keywords: - Surreal - Profound - Quiet - Premium - Modern - Thought-provoking - Cinematic - Visually shocking - Suitable for spreading ## VII. Render Quality - Ultra-high details - 8K texture - Realistic crowd simulation - Deep focus image, ensuring the entire crowd portrait is clearly visible - Rich details of miniature figures - Realistic spatial proportions - Clean picture, no messy background - Premium commercial visual texture - Similar to Octane Render / High-end 3D conceptual rendering effects ## VIII. Negative Prompts Please avoid the following issues: - Do not turn the character into a stranger - Do not just generate an ordinary portrait - Do not generate a simple collage - Do not let the crowd lose the portrait outline - Do not misplace facial features - Do not blur the facial structure - Do not use an exaggerated cartoon style - Do not make a low-resolution poster - Do not have a messy background - Do not make all the tiny people look exactly the same - Do not include text, watermarks, or logos - Do not over-beautify the character - Do not change the character's sense of age and real temperament