OpenZhenZhen AI Prompt Gallery · gpt-image-2.5 · image

Three-frame portrait in dim red light at a Japanese wooden inn

Three-frame portrait in dim red light at a Japanese wooden inn — explore this gpt-image-2.5 creation on OpenZhenZhen, with its original prompt and a reusable generation example.

Three-frame portrait in dim red light at a Japanese wooden inn — AI image creation
Three-frame portrait in dim red light at a Japanese wooden inn

Prompt

Chinese prompt

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。创作一张竖向、低分辨率的网络摄像头拼贴，恰好包含自上而下排列的三帧。三帧均为随附面部照片中的同一位成年女性。完整保留她的面部：五官比例、脸型、眼睛大小、鼻子、嘴巴、下颌线、年龄和性别均不得改变。保持原有发型，包括长度、分缝和质感；只将发丝呈现为自然略显凌乱、粗粝的状态。不要让她显得更年轻或过度可爱，也不要加入其他人物。 由于随附人物为女性，三帧均穿同一件带红色花卉图案的黑色浴衣。色彩与随附面部照片的色彩倾向一致，忠实呈现其色彩。背景为暗色、虚化、细节简洁的日式木质旅馆。 第1帧：呈现悲情电影主角般的悲伤神情，一根手指靠近嘴边。第2帧：头部略微转向一侧，展现侧脸；整体像随意自拍，不要过度摆拍。第3帧：露出温暖美丽的微笑；双手合十托住下巴，氛围平静而梦幻。 三帧统一采用 cinesteel 800T 胶片风格和电影光效。每帧一侧均施加相同的红色光线，并统一使用从左向右、雅致克制的运动模糊。始终保持同一人物身份和一致的视觉氛围。

English prompt

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 vertical, low-resolution webcam collage with exactly three frames stacked from top to bottom. Use the same single adult woman from the uploaded face photo in all three frames. Preserve her face exactly: facial proportions, face shape, eye size, nose, mouth, jawline, age, and gender must remain unchanged. Keep her original hairstyle, including its length, part, and texture; only render it naturally a little messy and rough. Do not make her younger or overly cute, and do not add any other person. Because the uploaded person is female, she wears the same black yukata with red floral patterns in all three frames. Use the same color palette as the uploaded face photo, keeping its colors faithful. The background is a dark, blurred Japanese wooden Ryokan with minimal detail. Frame 1: a sad expression like the protagonist of a sad film, with one finger near her mouth. Frame 2: her head turned slightly to the side to show her profile; casual selfie feeling, with no excessive posing. Frame 3: a warm, beautiful smile; both hands clasped together and supporting her chin, with a calm, dreamy mood. Keep the same cinesteel 800T film style and cinematic lighting in every frame. Apply the same red light from one side and tasteful left-to-right motion blur consistently across all three frames. Preserve one-person identity and visual atmosphere throughout.

View API model

Explore more AI image and video prompts · Try image generation · Try video generation · Create on the AI canvas