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

实景照片与荒诞手绘拼贴

Real Photos and Absurd Hand-Drawn Collage

将输入照片上半部保留为完整实景，下半部以同一场景为基础加入2–4处干墨、蜡笔和彩铅手绘介入，形成荒诞幽默的拼贴效果。

实景照片与荒诞手绘拼贴 — AI image creation
实景照片与荒诞手绘拼贴

提示词 / Prompt

中文提示词

以已配置角色参考ref-10a-v1的成年女性出现在画面中的原始照片作为视觉参考，创作纵向双区拼贴：上半部完整保留自然色彩的真实照片；下半部以同一场景为基础进行克制的混合媒材重构。上下两部分均保留原女性的身份、真实主体数量、姿态和空间关系。将2–4处荒诞手绘介入直接融入场景，包括一个夸张醒目的手绘角色，以及一个靠近边缘的小型幽默细节。可加入拟人化线条、故意错位的比例、小道具、箭头和一个写有短句哦！的气泡。使用干墨、蜡笔和彩色铅笔绘制，保留可见的手绘力度、纸张毛边、轻微渗墨和俏皮的拼接错位。保持原照片的自然色彩，仅在新增笔触中使用黑色、奶油色、朱红、钴蓝和芥末黄。下半部重构应让人一眼认出是同一场景。不要改变女性身份或姿态，不要添加额外的真实人物，也不要遮挡原图关键地标。

English prompt

Use a source photograph featuring the configured adult woman from character reference ref-10a-v1. Create a vertical two-part composition: retain the complete, natural-color real photograph in the upper half; in the lower half, reinterpret that same scene with a restrained mixed-media collage. Preserve the original woman's identity, number of real subjects, pose and spatial relationships across both halves. Add 2–4 absurd hand-drawn interventions directly into the scene, including one exaggerated, prominent doodle character and one small comic detail near an edge. The interventions may use anthropomorphic lines, deliberately mismatched proportions, small props, arrows and a speech bubble containing the short text Oh!. Draw them with dry ink, crayon and colored pencil, with visible hand pressure, paper-deckle edges, slight ink bleed and playful misregistration. Keep the source photograph's natural palette; confine added color to black, cream, vermilion, cobalt blue and mustard yellow strokes. The lower reinterpretation should remain immediately recognizable as the same scene. Do not alter the woman's identity or pose, add extra real people, or obscure key landmarks.

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

  • 1girl

    gpt image2.5生成

  • 法式风情地标美食海报

    A premium, clean and elegant poster concept built from the source’s broad request. Eiffel Tower, Arc de Triomphe, ratatouille and Provençal bouillabaisse are curator-selected illustrative examples, not details stated by the source; they can be replaced in a future variant. The source example image i

  • 梦回江户浮世绘古典人物像

    A reference-led Edo-period ukiyo-e portrait transformation that preserves the supplied person’s facial identity while restyling hair, clothing and accessories in traditional Japanese fashion. The current ref-10a-v1 is the generic person preview for this workflow, not a claim about any external named

  • 盛世长安万国来朝图

    A text-driven historical reconstruction of the source scene: an aerial view of Kaiyuan-era Chang’an, Emperor Xuanzong and Yang Guifei looking out from Hanyuan Hall, and a distant tribute procession. The configured generic portrait is used only as a face-identity anchor for Yang Guifei in this genera

  • 威震天高定西装时尚大片

    A text-driven fashion-cover portrayal of Megatron, retaining the source's silver-gray mecha, haute-couture suit and cyberpunk neon setting; it does not claim an exact adaptation-specific design. / 依据文字创作的威震天时尚封面，保留银灰机甲、高定西装与赛博朋克霓虹场景；不声称复现某一改编版本的精确造型。

  • 黑木崖红衣飞针傲立悬崖

    An original cinematic wuxia interpretation of the named fictional character, preserving the source's flying embroidery needles, red clothing, long hair, Heimuya cliff and blood-red sunset. The configured generic portrait is used only as the required face identity anchor, not as a canonical character