PROMPT RADAR / IMAGE视频叙事
社交媒体帖子 - Street Fashion Wall Sketch Collage
Creates a four-panel editorial fashion collage blending realistic street portraits with oversized hand-drawn wall-sketch alter egos.
@Alina_with_Ai 0 热度 0 浏览
ORIGINAL VISUAL / 来源见文末编辑先看
它为什么值得进入灵感库
它最有价值的部分,是把一个模糊想法推进到了可以被模型执行的程度。 把一段描述拆成起始状态、动作变化、镜头响应和结束画面,连续性比形容词更重要。
01 / 原始方法
完整视觉 Prompt
Goal: Create a square 2x2 editorial photo-collage where real street-style fashion portraits are combined with oversized hand-drawn black ink sketch versions of the same subjects on a plain beige/gray wall, like a creative “photo to doodle illustration” transformation showcase.
Canvas: Square 1:1 image divided into exactly 4 equal panels with thin white gutters between panels. Each panel has a minimal urban backdrop: smooth concrete wall, gray floor, soft natural light, muted beige and gray tones, realistic shadows, and editorial fashion photography quality.
Layout: Use exactly 4 panels:
1. Top-left panel: one stylish woman leaning against a beige wall on the left, wearing sunglasses, a loose white button-up shirt, blue jeans, white sneakers, and a black shoulder bag, holding a takeaway coffee. To her right, draw a large black pen sketch version of her sitting casually with one leg raised, holding a phone, with headphones around the neck. Add small doodled hearts, stars, and motion marks around the sketch. Place one small rounded-corner inset photo thumbnail near the bottom center-right with a white border.
2. Top-right panel: one woman in a white oversized shirt dress and white sneakers leaning on one leg beside a short black bollard, holding a magazine/newspaper in one hand and a takeaway coffee in the other, wearing sunglasses and a black shoulder bag. To her right, draw a large playful black ink sketch of the same woman standing on one leg with the other leg kicked back, holding a coffee and raising a magazine labeled “VOGUE.” Add small hearts, stars, and motion lines. Place one small rounded-corner inset photo thumbnail near the bottom left with a white border.
3. Bottom-left panel: two streetwear fashion subjects leaning against a wall, both wearing dark caps, sunglasses, oversized neutral jackets, dark baggy pants, and sneakers, each holding takeaway coffee. To their right, draw a large black ink sketch version of the same two people standing side by side in similar poses and outfits. Add small hearts, stars, and motion marks. Place one small rounded-corner inset photo thumbnail near the bottom left with a white border.
4. Bottom-right panel: one sporty woman standing on the left in a white crop top, loose white sweatpants, white sneakers, headphones, sunglasses, and a small black shoulder bag, looking down at a phone. To her right, draw a large black ink sketch version of her dancing or posing with one arm raised high and one leg bent, wearing the same headphones and outfit. Add small hearts, stars, and motion marks. Place one small rounded-corner inset photo thumbnail near the bottom left with a white border.
Subject style: Ultra-realistic editorial fashion photography for the real people, natural skin tones, accurate fabric texture, casual candid poses, modern street-style styling. The illustration portions should look like loose hand-drawn fashion sketches in black pencil/ink directly on the wall, with expressive linework, crosshatching, imperfect contours, and sketchy shading.
Text content: The only readable text should be the sketch magazine title {argument name="magazine title" default="VOGUE"} in the top-right panel. Avoid any other logos, captions, watermarks, or extra typography.
Customizable details: Keep the overall concept as {argument name="concept" default="real fashion portraits beside oversized wall-sketch alter egos"}; use {argument name="wall color" default="warm beige gray concrete"}; keep the clothing palette {argument name="clothing palette" default="white, denim, black, taupe, and gray neutrals"}; render in {argument name="visual style" default="ultra-realistic editorial photography mixed with loose black ink fashion illustration"}.
Constraints: Exactly 4 panels, exactly 4 inset thumbnails total, exactly 5 real photographed people total across the collage, exactly 5 large sketched people total across the collage. Maintain clean square collage composition, soft daylight, high-resolution detail, no extra panels, no frame around the whole image, no watermark.
02 / 复刻路径
不要一次改完所有变量
用 GPT Image 2 复刻时,先固定主体和构图,再逐次替换光线、色调或材质中的一个变量,这样更容易判断哪项描述真正起作用。
- 主体:Creates a four-panel editorial fashion collage blending realistic street portraits with oversized hand-drawn wall-sketch alter egos.
- 视觉方向:社区实测摄影提示词;按原提示词的光线描述执行
- 镜头与构图:按原提示词的镜头与景别执行;按原提示词的构图与画幅执行
- 交付约束:auto,遵循原提示词中的避免项;不要水印、乱码、低清、畸形手、错误透视或主体漂移。
负向约束
遵循原提示词中的避免项;不要水印、乱码、低清、畸形手、错误透视或主体漂移。
模型建议
GPT Image 2 / Grok Imagine / Midjourney / Flux。先用原始 Prompt 建立基准结果,再根据模型对自然语言、镜头运动或文字排版的能力做局部改写。
来源与编辑原则