OpenZhenZhen AI 프롬프트 갤러리 · gpt-image-2.5 · image

붉은 드레스 콜라주 인물의 정밀 얼굴 교체

Precise Face Replacement in a Collage Portrait of a Woman in a Red Dress

‘붉은 드레스 콜라주 인물의 정밀 얼굴 교체’은(는) gpt-image-2.5로 만든 작품입니다. OpenZhenZhen에서 원본 프롬프트와 생성 예시를 살펴보고 새로운 창작에 활용해 보세요.

붉은 드레스 콜라주 인물의 정밀 얼굴 교체 — AI image creation
붉은 드레스 콜라주 인물의 정밀 얼굴 교체

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。 将上传的来源图像作为精确的“目标图像”。只更改目标图中女性的面部身份:替换为已配置的参考身份 ref-10a-v1,并将这一身份一致地应用于中央女性及周围四张较小的相框照片。除面部身份外,目标图像是所有内容的唯一底图。 严格保留目标图像原有的发型、发长、发色与头发位置、耳环、太阳镜及其准确位置、红色连衣裙、体型、手部、姿势、包、鞋子、灯光、阴影、相机角度、透视、背景、花朵、相框、纸张纹理、装饰、手写文字、爱心和整体构图。不要复制已配置参考人物的发型、服装、身体、姿势、背景或配饰。保留目标图中每处原有表情、头部角度、视线方向和姿势;让同一参考面部自然适配不同视角,不改变这些细节。 面部身份替换应无缝、写实、高分辨率且具有摄影质感。不得添加或删除任何物体或文字。不要进行其他编辑,不要增加人物、改动原有文字、更改太阳镜、生成乱码或水印。

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. Use the uploaded source image as the exact TARGET IMAGE. Change only the facial identity of the woman in that target: replace it with the configured reference identity, ref-10a-v1. Apply that same identity consistently to the central woman and all four smaller framed photos around her. Treat the target image as the master for everything except facial identity. Preserve the target image's original hairstyle, hair length, hair color and placement, earrings, sunglasses and their exact position, red dress, body shape, hands, poses, bag, shoes, lighting, shadows, camera angle, perspective, background, flowers, photo frames, paper textures, decorations, handwritten text, hearts and entire composition exactly. Do not copy the configured reference person's hairstyle, clothes, body, pose, background or accessories. Keep each original expression, head angle, gaze direction and pose; adapt the same reference face naturally to each view without changing those details. Make the identity replacement seamless, realistic, high-resolution and photorealistic. Add or remove no object or text. No other edits, extra people, altered wording, changed sunglasses, garbled text or watermark.

API 모델 보기

더 많은 이미지·동영상 프롬프트 보기 · 이미지 생성 · 동영상 생성 · AI 캔버스에서 만들기