Thanks to visit codestin.com
Credit goes to promptsref.com

图片转提示词生成器

上传参考图,生成可编辑、可复制、可复用的提示词

上传图片
上传图片,或从你的素材中选择
点击上传或选择图片
PNG / JPG / WEBP
上传或选择图片
结果
输出将显示在这里
等待中
使用说明
上传或选择图片后,点击 分析
提示
  • 分析过程中请勿刷新页面。
  • 不允许 NSFW 内容。提交此类内容可能失败,且积分不予退还。
  • 所选图片仅用于本次解析。

它会从图片里读出什么?

它不只看主体,也会提取风格、光线、构图、镜头感和影响画面质感的小细节,方便你更接近原图效果。

主体和场景

识别主要人物、物体、动作、环境,以及它们之间的关系。

风格和媒介

判断画面更接近摄影、电影感、插画、3D、动漫、商品图还是海报。

光线和色彩

提取光源方向、明暗对比、色彩组合、阴影和整体情绪。

构图和镜头

把取景、视角、镜头感、景深、透视和裁切方式写进提示词。

细节和氛围

补充材质、服装、纹理、环境、情绪和画面完成度。

可编辑提示词

输出可继续修改的提示词,可用于 Midjourney、GPT image、Nano Banana 等模型。

Promptsref 优势

不只是生成一句图片描述

我们会把图片当成一份提示词说明来解析,同时给出详细的 JSON 提示词和更短的自然语言提示词。

01

结构化 JSON 提示词

JSON 版本会拆出主体、姿势、服装、光线、镜头、背景、氛围、限制条件和反向提示词。信息越清楚,模型越容易复刻原图的视觉效果。

02

同时给出短提示词

除了详细 JSON,我们也会给一版自然语言提示词。它更短,更适合快速复制、轻量修改和直接生成。

03

支持多语言输出

解析前可以选择输出语言。图片细节会保留下来,但提示词会用你习惯的语言表达。

图片转提示词示例

下面这些示例展示图片如何对应到可复用、可修改的提示词。

由提示词生成的街头摄影 AI 图片示例

街头摄影参考

保留姿势、低角度、阳光和手机随拍感。

提示词

Street photography, sunlight, trendy Korean gril, 20s, squatting on the ground, strong perspective, low angle, shot with iPhone 17.

由提示词生成的杂志封面人像 AI 图片示例

杂志封面人像

保留面部、妆容、道具、排版、光线和封面氛围。

提示词

{ "subject": { "description": "A young woman in an extreme close-up editorial cover portrait, her face filling almost the entire frame while a long snack stick rests against her lips.", "mirror_rules": "not applicable; this is a direct-camera editorial cover portrait, not a mirror selfie", "age": "early 20s", "expression": { "eyes": { "look": "soft but piercing gaze, slightly sleepy, directly engaging the viewer", "energy": "low-motion, magnetic, detached", "direction": "straight into camera from a slightly lowered head" }, "mouth": { "position": "slightly parted lips touching a long pale snack stick", "energy": "gentle tension, absent-minded, intimate" }, "overall": "dreamy, aloof, quietly provocative, emotionally withheld" }, "face": { "preserve_original": "keep the original facial proportions: large eyes, small nose, soft narrow jawline, short philtrum, full blurred lips", "makeup": "soft shimmery lids, subtle eyeliner, bright under-eye area, lightly defined lashes, peach-coral gradient lips, minimal contour" } }, "hair": { "color": "deep black", "style": "medium-length loose hair with wispy uneven bangs falling over the forehead and grazing the eyes", "effect": "slightly messy strands catch the light and create a soft veil around the face" }, "body": { "frame": "slim upper body with narrow shoulders and delicate collarbones", "waist": "not visible", "chest": "upper chest and cleavage softly visible through a deep neckline", "legs": "not visible", "skin": { "visible_areas": "face, neck, collarbones, upper chest, shoulder, fingers", "tone": "pale ivory with a soft warm undertone", "texture": "creamy, velvet-soft, pore-blurred by bloom rather than retouching", "lighting_effect": "strong frontal light turns the skin milky and luminous, with clipped highlights and a gentle halo" } }, "pose": { "position": "very close to camera, head slightly lowered, shoulders angled softly", "base": "one hand lifts a long pale snack stick to the lips while the face fills nearly the entire frame", "overall": "an intimate pause captured at almost uncomfortable closeness" }, "clothing": { "top": { "type": "thin-strap glossy camisole or slip-style top", "color": "black", "details": "plunging neckline, glossy surface, minimal visible structure", "effect": "catches light like coated satin or patent fabric" }, "bottom": { "type": "not visible", "color": "not visible", "details": "out of frame" } }, "accessories": { "jewelry": "a delicate silver-toned necklace with a small sparkling pendant", "prop": "a long pale snack stick or biscuit held to the mouth" }, "photography": { "camera_style": "fashion magazine cover portrait with social-media intimacy", "angle": "extreme close-up from near eye level with a slight downward tilt and mild asymmetry", "shot_type": "head-and-upper-chest close-up, cover portrait", "aspect_ratio": "3:4 vertical", "texture": "soft-focus bloom, glossy print feel, mild haze, low-contrast highlights, subtle digital softness", "lighting": "hard frontal light diffused by lens bloom or haze, creating glowing skin and a shadow pocket under the chin", "depth_of_field": "shallow-to-medium focus, with facial features readable but edges melting softly" }, "background": { "setting": "the background largely disappears behind hair, skin glow, and editorial typography", "wall_color": "not clearly visible", "elements": [ "oversized white magazine masthead \"SPICY\" across the top", "small white issue date in the upper left", "stacked white cover lines on the right side", "large white bottom headline anchored around the word \"WINTER\"", "barcode block in the lower left corner" ], "atmosphere": "glossy, dreamy, collectible, like a pop-fashion issue balanced between intimacy and performance", "lighting": "the scene is visually dominated by face-lit highlights while the surroundings stay swallowed by darkness and typography" }, "the_vibe": { "energy": "quiet, magnetic, low-temperature intensity", "mood": "dreamy, detached, intimate, fashion-editorial", "aesthetic": "soft-bloom pop cover, slightly Y2K, glossy print sensuality", "authenticity": "keep the tiny imperfections alive: uneven bangs, slight overexposure, off-center framing, real hand tension, accidental-looking timing", "intimacy": "the camera sits so close it feels like a private second that was made public", "story": "She had only paused for one distracted bite, and the lens turned that unfinished moment into a cover image people would keep.", "caption_energy": "teasing, collectible, cool-girl cover-line confidence" }, "constraints": { "must_keep": [ "extreme close-up composition with the face dominating the frame", "deep black wispy bangs crossing the forehead and partly shading the eyes", "long pale snack stick touching the lower lip", "milky overexposed skin glow with dreamy bloom", "black glossy thin-strap neckline visible at the bottom", "bold white magazine typography and barcode layout", "readable clean Latin-letter cover design rather than random gibberish" ], "avoid": [ "mirror selfie cues", "busy room background", "wide smile or cheerful expression", "over-sharpened skin", "heavy contour or dramatic glam makeup", "extra fingers or deformed hand", "multiple props", "full-body framing", "flat matte lighting", "cartoonish stylization" ] }, "negative_prompt": [ "mirror selfie", "busy interior", "wide grin", "extra fingers", "deformed hand", "multiple snack sticks", "heavy contour", "thick eyeliner", "over-sharpened skin", "matte lighting", "full body shot", "wide-angle distortion", "clean studio backdrop", "neon colors", "text gibberish", "missing barcode", "anime style" ] }

由提示词生成的动漫主视觉 AI 图片示例

动漫主视觉

保留角色站位、俯拍构图、图形文字和赛璐璐风格。

提示词

Five handsome characters from Jujutsu Kaisen — Satoru Gojo, Suguru Geto, Megumi Fushiguro, Yuji Itadori, and Kento Nanami — standing at the center of a vast black-and-white typography ground, high-angle wide overhead composition, loose theatrical group staging with strong depth layers. Foreground lead: Gojo in dark high-collar uniform and blindfold or sunglasses, white hair as the visual anchor; others in dark jujutsu sorcerer uniforms with distinct silhouettes (Geto’s long hair and composed menace, Megumi’s spiky black hair, Yuji’s athletic build, Nanami’s suit-and-tie sharpness). Premium Japanese anime key visual and visual-novel cover quality. Outfits: high-contrast dark uniforms, collars, coats, ties, belts; hems expanding in perspective. Some hold weapons or cursed-energy gesture poses, frozen post-movement before the next fight. No smiles; calm, restrained, slightly distant gazes toward the camera area. White brick ground with giant black English typography, near-noon hard directional light, crisp shadows, high graphic contrast. Anime cel shading, sharp lineart, slight perspective exaggeration, subtly off-center composition.

生成后可用于不同模型

保留视觉描述,再按你使用的模型补充比例、参数或风格语法。

打开 AI 图片生成器

GPT image 2.0

适合用自然语言描述主体、姿势、镜头、光线和画面质感。

Nano banana

适合参考图复刻、视觉改写,以及需要理解图片细节的场景。

Midjourney

适合风格复刻、电影感参考、SREF 探索和概念图生成。

其他 AI 生成模型

把提取出的提示词作为基础,再补充对应模型的参数或语法。

什么时候适合用图片转提示词

拆解一张参考图的风格和画面结构。

把人像、商品图、海报或插画整理成可复用提示词。

把视觉灵感转换成 GPT image、Nano Banana、Midjourney 等模型可用的 prompt。

学习高质量图片提示词如何描述主体、风格、光线和构图。

在保持同一视觉方向的前提下,扩展更多 prompt 变体。

怎么使用图片转提示词

用于转换成 AI 图片提示词的示例图片

提示词

这张图片是一座天使雕像坐在白色花丛中的照片。天使穿着白色长袍,背后有天使翅膀。她手中拿着一朵小橙色花,似乎正在闻它的香气。背景是模糊的,但看起来是一片田野,上方是清澈的蓝天。整体氛围宁静祥和。

📱

上传或选择要分析的图片

等待几秒钟生成提示词

😊

复制提示词,或先修改再使用