OpenZhenZhen AI 提示词画廊 · gpt-image-2.5 · image
明媚日光重塑光影,人物构图完美保留
Bright sunlight reshapes light and shadow, perfectly preserving the figure composition
A controlled image-edit template that relights an existing image as bright exterior daytime while preserving the people, objects, identities, poses, expressions, clothing, setting geometry, camera angle, framing, and composition exactly. Only illumination, exposure, shadows, reflections, and dayligh

提示词 / Prompt
中文提示词
重新打光提供的图像,使其呈现明亮的室外白天效果。保持人物完全不变:保留每个人的身份、面部、发型、肤色、服装、表情、姿势、动作和身体比例。保持物体、环境、相机角度、裁切、透视、空间深度关系和相同构图完全不变。彻底改变光照且只改变光照:用可信的明亮户外日光、自然天空补光、物理一致的方向与柔和度、日间曝光、真实接触阴影、表面反射和符合现实的日光色彩响应替换原有照明。保留原图所有细节和空间关系;不要添加、删除、移动、重新造型或重新解释任何内容。结果应像同一张照片在明亮白天的室外拍摄,而不是重新构图的新图像。避免身份漂移、解剖结构改变、服装变化、新道具、虚假HDR光晕、塑料感皮肤、不一致阴影、文字、水印和标志。
English prompt
Relight the provided image to create a bright exterior daytime scene. Keep the people exactly the same: preserve every identity, face, hairstyle, skin tone, clothing, expression, pose, gesture, and body proportion. Keep the people unchanged and keep the objects, environment, camera angle, crop, perspective, depth relationships, and same composition exactly the same. The edit must keep the people unchanged. The instruction is to change the lighting entirely and only: replace the original illumination with believable bright outdoor daylight, natural sky fill, physically consistent direction and softness, daylight exposure, realistic contact shadows, surface reflections, and true-to-life daylight color response. Preserve all original details and spatial relationships; do not add, remove, move, restyle, or reinterpret anything. The result should look like the same photograph captured outdoors in bright daytime, not a newly composed image. Avoid identity drift, altered anatomy, changed clothing, new props, artificial HDR halos, plastic skin, inconsistent shadows, text, watermarks, and logos.
探索更多图片和视频提示词 · 生成图片 · 生成视频 · 在 AI 画布中创作
相关作品
- AI智能换装试穿对比海报
A four-section 16:9 fashion try-on and pose showcase, using the configured generic identity reference, a coordinated outfit flatlay, a matching try-on result, and six poses.
- 20 岁希娜时尚个性海报
A vertical 9:16 Y2K street-fashion character poster using the configured generic face-identity input, with an adult Sheena design, irregular detail panels, and exact source metrics text.
- 3×3 九宫格拼贴写真
A cohesive 3×3 Korean playful studio portrait collage preserving the visible identity in the configured person reference across nine distinct poses and expressions.
- 蓝莓主题可爱风人物海报
A centered 3:4 blueberry-themed poster portrait using the configured generic face-identity reference, plush blueberry props, and exact source lettering.
- 韩风手绘涂鸦感人物海报
A Korean-style illustrated half-body portrait poster that preserves the visible identity in the configured person reference, with expressive ink-and-pastel doodle art.
- 几何折纸蓝图风蜘蛛侠
Hyper-detailed folded-paper SpiderMan blueprint preserving the source geometry, blueprint annotations, minimal paper-and-ink palette, studio lighting, crisp shadows, architectural-art style and photorealistic paper detail. The configured reference is attached only as the workflow-required generic id