OpenZhenZhen AI 프롬프트 갤러리 · gpt-image-2.5 · image
타임스퀘어의 시공간 교차
Convergence of Time and Space in Times Square
‘타임스퀘어의 시공간 교차’은(는) gpt-image-2.5로 만든 작품입니다. OpenZhenZhen에서 원본 프롬프트와 생성 예시를 살펴보고 새로운 창작에 활용해 보세요.
프롬프트
중국어 프롬프트
以用户上传的参考图作为主要可见人物的面部身份依据。在不与身份参考冲突的情况下，保留来源的场景、造型、服装、姿势、构图、光线和视觉风格。 创作一张照片级写实的横向4:3时空穿越电影场景，地点为纽约时代广场；默认：纽约时代广场。左侧呈现纽约1920年代；默认：纽约1920年代。右侧呈现纽约当代；默认：纽约当代。城市景观从中央自然连续地演变，不出现硬切分界。左侧使用具有时代感的剧院立面、早期电气招牌、当年店面、复古汽车和符合年代的街景细节；右侧呈现玻璃钢结构建筑、大型电子广告牌、现代出租车和当代街道。两侧保持同一处可辨认的路口与透视关系。画面中只有一位明确成年的女性，但以同一人物在两个时空中的两个版本呈现：左侧穿着优雅的1920年代服装，右侧穿当代城市时尚服饰；两个面孔都忠实绑定固定身份参考ref-10a-v1。让两个版本隔着无缝过渡区域彼此注意到对方，以自然惊讶的表情和指向现代物件的细微手势呈现互动；人体结构与尺度真实可信。让路面、立面、光线和街道道具在中央平滑转换，使两个时代真实交融，而不是简单并排对照。采用广角街道视角、电影感自然光、丰富建筑纹理和克制的空气纵深。保留清楚的建筑结构，不依赖精确可读的店招文字或真实标志。不要额外人物、硬切分界、重复多余肢体、历史一侧出现当代屏幕、8K承诺、叠加文字、标志或水印。
영어 프롬프트
Use the uploaded reference image as the facial identity reference for the main visible person. Keep the source scene, styling, clothing, pose, composition, lighting and visual style unless they conflict with the identity reference. Create a photorealistic horizontal 4:3 cinematic time-travel scene at Times Square, New York; default: Times Square, New York. Show 1920s New York on the left; default: 1920s New York. Show present-day New York on the right; default: present-day New York. Let the city evolve continuously through the center with no hard dividing line. On the left, use historically evocative theater facades, early electric signs, period storefronts, vintage automobiles and era-appropriate street details; on the right, show contemporary glass-and-steel architecture, large electronic billboards, modern taxis and current streetscape. Keep the same recognizable intersection and perspective across both halves. Depict ONE clearly adult woman twice as the same person at two points in time: one version in refined 1920s clothing on the left and her present-day version in contemporary city fashion on the right. Bind both faces to fixed identity reference ref-10a-v1. Have the two versions notice one another across the seamless transition, with a natural surprised expression and a subtle gesture toward a modern object; keep anatomy and scale realistic. Blend pavement, facades, lighting and street props smoothly through the center to make the eras interact rather than read as a simple side-by-side comparison. Use a wide-angle street-level view, cinematic natural light, rich architectural texture and restrained atmospheric depth. Preserve legible structure without relying on exact readable storefront copy or real logos. No additional people, hard split seam, duplicated extra limbs, present-day screens on the historical side, 8K claim, text overlay, logo or watermark.
더 많은 이미지·동영상 프롬프트 보기 · 이미지 생성 · 동영상 생성 · AI 캔버스에서 만들기