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

실사 인물 사진과 빈티지 수채화 편집 일러스트 상하 분할 포스터

Real Portrait and Vintage Watercolor Editorial Illustration Split Poster

‘실사 인물 사진과 빈티지 수채화 편집 일러스트 상하 분할 포스터’은(는) gpt-image-2.5로 만든 작품입니다. OpenZhenZhen에서 원본 프롬프트와 생성 예시를 살펴보고 새로운 창작에 활용해 보세요.

실사 인물 사진과 빈티지 수채화 편집 일러스트 상하 분할 포스터 — AI image creation
실사 인물 사진과 빈티지 수채화 편집 일러스트 상하 분할 포스터

프롬프트

중국어 프롬프트

创作一张竖版上下双拼海报，清晰区分上下区域，留白充足，整体轻松、俏皮、温柔、时髦、复古，并带少量幽默感。将完整的ref-10a-v1照片作为唯一整图输入。上半部分保留真实摄影风格，完整保留整张照片，不裁切、不变形，包括深青蓝背景、原有取景与光线，以及成年女性的直视目光、棕色中分长发、黑色细肩带上衣、肩部、姿势与平静表情。将整幅横向照片按比例完整放入上半区域；如需适配画布可添加留白，不要重绘或修饰照片。下半部分把同一位成年女性转译为复古水彩手绘编辑插画，保留可辨认的头肩轮廓、姿势、发型与衣着。以概括表达而非机械复制细节，运用简练形体、略带迟疑的手绘轮廓、适度夸张比例和少量俏皮视觉隐喻，同时保持身份可辨。融合现代主义编辑插画、包豪斯平面构成、儿童绘本、稚拙艺术与时尚速写气质。配色从照片实际可见颜色中提取，采用高明度、低至中等饱和度的柔和有限色盘，以近似色统一氛围，只用少量有照片依据的互补色作跳点。混合水彩、水粉、彩铅、粉蜡笔与干刷质感，保留纸张颗粒、透底、笔触断裂、边缘毛糙、轻微脏色与细微旧印刷套色偏差。插画背景使用米白或浅色并大面积留白，仅以松散淡彩、局部斜线或简单几何记号承托人物。让人物特征、色彩、姿势与视觉符号呼应照片。避免复杂场景堆积、插画中的写实光影、光滑矢量、三维渲染与过度精修。中文输出时仅使用一个简短、非事实性标题“静谧肖像”和两组极小辅助文字：编号“01”与说明“柔光，静望”。标题采用略带稚拙复古感的手写艺术字，辅助文字使用克制温和的无衬线或衬线字形。文字可轻微倾斜、错位、拉伸或产生套色偏移，让它自然融入插画而非后贴排版。只使用照片中可见的内容与情绪，不虚构地点、日期、事件、身份或背景故事。不要添加另一语言的标题、第二个标题或额外文字组。保留轻松、聪明、幽默、浪漫、略显笨拙却古怪自信的气质，像一页被精心保存的旧时尚绘本与现代艺术杂志。

영어 프롬프트

Create a vertical top-and-bottom diptych poster, clearly separating upper and lower panels, with generous whitespace and a relaxed, playful, gentle, fashionable retro editorial mood with a touch of humor. Use the complete ref-10a-v1 image as the sole full-photo input. In the upper panel, preserve the whole photograph uncropped and undistorted in its real photographic style, including the deep teal-blue background, original framing and light, and the adult woman's direct gaze, long center-parted brown hair, black thin-strap top, shoulders, pose and calm expression. Fit the complete landscape image proportionally inside the upper panel, adding margin if needed; do not redraw or retouch it. In the lower panel, translate the same adult woman, recognizable head-and-shoulder outline, pose, hair and clothing into a retro watercolor hand-drawn editorial illustration. Re-express rather than mechanically copy details, with concise shapes, slightly hesitant handmade outlines, gently exaggerated proportions and a small playful visual metaphor while keeping her identity recognizable. Blend modernist editorial illustration, Bauhaus graphic composition, children's picture-book, naive-art and fashion-sketch sensibilities. Derive a high-brightness, low-to-medium-saturation soft limited palette from colors actually visible in the portrait, using analogous hues for unity and only small source-grounded complementary accents. Mix watercolor, gouache, colored pencils, pastels and dry brushes; retain paper grain, transparency, broken brushstrokes, rough edges, slight pigment muddiness and subtle old-print color-registration shifts. Give the illustration an off-white or light field with large areas of white space; support the figure only with loose light color, a local diagonal or a simple geometric mark. Keep subject features, colors, pose and visual symbols echoing the photo. Avoid scene clutter, realistic lighting or shadows in the illustration, smooth vectors, 3D rendering and over-refinement. For an English-language render, use one brief nonfactual title, “Quietly, in Color”, and only two small auxiliary groups: folio “01” and caption “Soft light, still gaze”. Set the title in naive retro handwritten art lettering and the auxiliary text in restrained gentle sans-serif or serif type. A slight tilt, offset, stretch or color-registration shift is allowed; let text grow naturally with the illustration rather than appear pasted on. Use only details and mood visible in the portrait; invent no location, date, event, identity or backstory. Do not add a second-language title, second title or extra text groups. Preserve a relaxed, smart, humorous, romantic, slightly awkward and quirky-confidence feeling, like a carefully kept old fashion picture-book page in a modern art magazine.

API 모델 보기

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