OpenZhenZhen AI 프롬프트 갤러리 · gpt-image-2.5 · image
도쿄 심야 플래시 아래 성인 여성의 모션 블러 장면
Adult Woman in Motion Blur Under Tokyo Midnight Flash
‘도쿄 심야 플래시 아래 성인 여성의 모션 블러 장면’은(는) gpt-image-2.5로 만든 작품입니다. OpenZhenZhen에서 원본 프롬프트와 생성 예시를 살펴보고 새로운 창작에 활용해 보세요.
프롬프트
중국어 프롬프트
创作一张 9:16 竖幅、超写实抓拍风格的智能手机照片，表现配置身份参考中的成年女性置身于摆拍的东京夜景。使用配置的身份参考保留她的面部与发型身份，不复制任何来源示例中的人物。场景设在安静的城市人行道上，旁边是一栋白色瓷砖外墙建筑。表现她以半侧身姿态走过镜头，在迈步时短暂回头，露出羞怯又略带俏皮的转瞬笑容。一只手抬到镜头附近，部分遮挡相机；身体继续向前，逐渐靠近画面边缘。构图保持不完美，仿佛路过时匆忙拍下。使用现代智能手机手持快速拍摄，并加入明显、失控的相机抖动。抬起的手、头发和面部带有强烈定向运动模糊，移动身体周围出现局部重影；面部表情因模糊而柔化，但整体人物仍可辨认。使用运动中触发的纯闪光灯照明：皮肤和抬起的手上有刺眼高光，周围深暗，曝光不均，局部高光溢出并形成硬阴影。让白色瓷砖墙延伸成条纹光带，街道环境退入黑暗，尽量不出现清晰可读的细节。整体感觉原始、短暂、羞怯而俏皮，带有极重噪点和明显模糊；避免电影感、时尚大片感或精修质感。避免窥视式构图、复制其他真实人物面孔、身份漂移、美颜滤镜、面部重塑、夸张曲线、人体变形、多余或缺失手指、标志、文字和水印。
영어 프롬프트
Create a 9:16 vertical, ultra-realistic candid-style smartphone photograph of the configured adult woman in a staged Tokyo night scene. Use the configured identity reference to preserve her face and hair identity; do not copy the person from any source example. Set the scene on a quiet city sidewalk beside a building with white ceramic-tile walls. Show her walking past the camera in half profile, turning her head back mid-step with a fleeting, shy and slightly playful smile. Raise one hand close to the lens so it partly blocks the camera, while her body continues forward and drifts toward the edge of the frame. Keep the framing imperfect, as if captured quickly while passing. Use a modern smartphone, handheld rushed capture and strong uncontrolled camera shake. Add heavy directional motion blur across the raised hand, hair and face, with partial ghosting around the moving body; keep the expression softly blurred but the overall figure recognizable. Use flash-only lighting fired mid-motion: harsh highlights on skin and the raised hand, deep surrounding darkness, uneven exposure, some blown highlights and hard shadows. Let the white tile wall stretch into streaked light bands and let the street fade into darkness with minimal readable detail. Keep the feeling raw, fleeting, shy and playful, with extreme image noise and heavy blur; avoid a cinematic, editorial or polished finish. Avoid voyeuristic framing, copying another real person's face, identity drift, beauty filters, face reshaping, exaggerated curves, distorted anatomy, extra or missing fingers, logos, text and watermarks.
더 많은 이미지·동영상 프롬프트 보기 · 이미지 생성 · 동영상 생성 · AI 캔버스에서 만들기