OpenZhenZhen AI 提示词画廊 · gpt-image-2.5 · image

虚实交织的撕裂边缘分裂肖像

Fractured Portrait at the Edge of Interwoven Realities

以方形居中近景呈现身份参考中的成年女性面部与上肩：左半边为线条干净、平涂配色和简化阴影的2D漫画插画，右半边为具备自然肤质与影棚光影的写实电影肖像。沿面部正中垂直对齐两种风格，并以窄幅撕裂漫画纸边缘表现从插画到现实的过渡。

虚实交织的撕裂边缘分裂肖像 — AI image creation
虚实交织的撕裂边缘分裂肖像

提示词 / Prompt

中文提示词

创作一张1:1方形分裂风格肖像，主体是身份参考ref-10a-v1中明确成年的女性；保持她可辨识的面部身份、比例、神情与五官位置。采用居中对称的近景，重点呈现完整面部和上肩，仅微妙露出部分服装。沿面部正中央垂直分开：左半边渲染为精致的2D漫画插画，线条干净有力、色彩平涂、阴影简化为赛璐璐风格；右半边渲染为写实电影感影棚肖像，呈现自然肤质、逼真细节与柔和真实光线。两半的眼睛、鼻子、嘴、下颌线和头部轮廓沿中心线对齐，组合成一张连贯的脸，而不是错位的两张脸。用一条窄而不规则的漫画纸撕裂边缘沿中央纵向分隔两种风格，且不得遮挡关键五官。使用柔和中性的影棚背景，以及戏剧性但平衡、在两侧都可信的灯光。不要额外面孔、人物、文字、标志、水印或其他分割线。保持方形画幅；不要承诺固定像素分辨率。

English prompt

Create one square 1:1 split-style portrait of the clearly adult woman represented by identity reference ref-10a-v1; preserve her recognizable facial identity, proportions, expression, and alignment. Use a centered, symmetrical close crop focused on her full face and upper shoulders, with only a subtle glimpse of her outfit. Divide the image vertically through the exact center of her face: render the left half as a polished 2D comic illustration with clean confident linework, flat color, and simplified cel shading; render the right half as a photoreal cinematic studio portrait with natural skin texture, lifelike detail, and soft realistic light. Align the eyes, nose, mouth, jawline, and head contours precisely across the center seam so both halves form one coherent face, not two offset faces. Separate the treatments with a narrow irregular torn-comic-paper edge running vertically down the center; keep it from obscuring key facial features. Use a soft neutral studio background and dramatic yet balanced lighting that remains credible across both halves. No extra faces, people, text, logos, watermark, or additional split lines. Square framing only; do not promise a fixed pixel resolution.

查看 API 模型

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