GPT Image 2AI 提示词
Japanese Street Style Lookbook — Cinematic Urban Portrait Prompt
Generate moody, cinematic street fashion portraits with handwritten Japanese text overlays, perfect for brand campaigns and social media visuals.

提示词 · EN
一张充满情绪的都市时尚大片,拍摄了一位年轻女性蹲在狭窄巷弄中,身旁是布满黑色涂鸦和贴纸的斑驳水泥墙,背景中的金属卷帘门上也布满了涂鸦。她穿着一套宽松的深蓝色运动套装,包含 2 件可见单品:拉链运动外套和配套运动裤,内搭一件浅灰色或米白色的连帽卫衣,脚踩白色厚底运动鞋。她扎着凌乱的丸子头,单手撑膝,另一只手靠近脸部,呈现出随性且内敛的街头风格。面部区域被一个矩形模糊块遮挡。构图为竖向,全身至四分之三身位,略微低角度拍摄,采用柔和的自然阴天光线,色调沉稳且具有电影感,浅景深,摄影风格写实,服装褶皱、路面纹理及墙面污垢细节丰富。在图像中添加 2 处独特的文字叠加:左上方为手写白色日文毛笔字,内容为 arghandwritten Japanese text: きゅっくらい、ちゃんと わたしを ほめてあげよ。;左下方为 5 行排列整齐的白色优雅衬线体编辑文字,内容为 argeditorial caption: Street is\nMy Everyday.\n2024\nAutumn / Winter\nLook Book,下方另有一行较小的第 6 行署名 argcredit text: by mellovneon。强调日系街头风 Lookbook 美学,风格随性而精致,略带粗粝感,具有高端时尚大片的质感。
关于这个提示词
Generate moody, cinematic street fashion portraits with handwritten Japanese text overlays, perfect for brand campaigns and social media visuals. 它更适合作为 GPT Image 2 的概念艺术起点:先保留画面结构、主体关系和风格约束,再替换成你的品牌、人物或场景。
使用时优先替换 handwritten Japanese text、editorial caption、credit text 等占位符,并保持镜头、构图和材质描述的顺序不变。这样更容易获得稳定的画面结果,也方便继续做多版本对比。
如何使用这个提示词
- Copy the prompt: Select and copy the entire prompt text provided.
- Choose your model: Use a high-quality image generation model like Sora 2 or similar AI that supports detailed text and aesthetic instructions.
- Tweak variables: Edit the
{argument}placeholders to customize the Japanese handwriting, editorial caption, and credit text to match your brand. - Adjust details: Modify outfit colors, location specifics, or lighting mood to fit your creative vision.
- Generate and refine: Run the prompt, review the output, and iterate by tweaking details until you achieve the desired cinematic look.



