PROMPT RADAR / PROMPT视频叙事
个人资料 / 头像 - Candid Motion Blur Portrait
An advanced prompt focusing on motion blur and realistic movement for candid photography, maintaining subject identity from a reference image.
@selinatasnim1 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
Preserve the subject’s identity and appearance with maximum accuracy. Do not change or reinterpret her facial features, facial proportions, eye shape, nose, lips, jawline, bone structure, skin texture, hair color, hair length, hair density, or natural hair texture. Maintain a highly faithful resemblance to the reference image throughout the scene.
Create a photorealistic candid photograph captured in the middle of natural movement. Use a close-up, waist-up composition. The woman is standing with her back partially toward the camera at the exact moment she quickly turns her head over her shoulder. Her gaze is directed naturally toward something off to the side, as if she has just noticed something behind her. Her expression should feel spontaneous and authentic, not posed or intentionally modeled for the camera.
Natural Motion & Sharpness
The sudden head movement creates realistic, subtle motion blur. Apply gentle blur across the moving face, hair, and individual curls, with slightly softened silhouette edges caused by the speed of the movement. Keep the face sufficiently recognizable and preserve the subject’s defining facial characteristics.
Her curls naturally sweep outward with the movement, creating a believable dynamic trail. The motion should feel like a genuine photograph captured by a real camera with a slightly slower shutter speed—not an artificial blur effect.
Body Language & Composition
Maintain a relaxed, natural posture with subtle asymmetry. The shoulders, neck, and upper body should show believable movement caused by the quick head turn. Avoid fashion-model posing, exaggerated gestures, or stiff studio positioning.
Lighting & Environment
Use cinematic directional lighting with soft, realistic shadows and atmospheric depth. Place the subject in a textured, believable environment with natural visual details. Preserve realistic skin with visible pores, fine texture, and subtle natural imperfections rather than overly smooth or plastic-looking skin.
Camera & Film Aesthetic
Make the image feel as though it was captured on a real 35mm film camera. Include pronounced but natural film grain, subtle optical imperfections, gentle lens characteristics, realistic depth of field, and authentic photographic texture.
Use a muted, low-saturation cinematic color palette with natural tones and restrained contrast. The final image should look like a genuine candid photograph taken in real life—organic, imperfect, cinematic, and completely photorealistic.
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:An advanced prompt focusing on motion blur and realistic movement for candid photography, maintaining subject identity from a reference image.
- 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