Thư viện prompt AI OpenZhenZhen · gpt-image-2.5 · image

Áp phích tự sự đôi bên với chân dung hiện thực phong cách dân gian ngây thơ

Naive Folk Art Portrait Split-Narrative Poster

Áp phích tự sự đôi bên với chân dung hiện thực phong cách dân gian ngây thơ — khám phá tác phẩm được tạo bằng gpt-image-2.5 trên OpenZhenZhen, kèm prompt gốc và ví dụ tạo nội dung để tham khảo.

Áp phích tự sự đôi bên với chân dung hiện thực phong cách dân gian ngây thơ — AI image creation
Áp phích tự sự đôi bên với chân dung hiện thực phong cách dân gian ngây thơ

Prompt

Prompt tiếng Trung

将每一张照片制作成一张独立的高级设计海报,不做多图拼接。将完整的ref-10a-v1照片作为唯一整图输入与身份参考。遵循来源末尾的最终更正:采用7:6横版左右分屏布局,严格左右1:1分割,左右各占画面50%,绝不使用上下布局。实际照片中是一位成年女性,棕色中分长发、黑色细肩带上衣、平静直视目光,背景为深青蓝色。左半保留原照片主体身份、结构、姿态、真实质感、自然光影和色彩氛围,仅作轻微艺术杂志式调色。适配左侧区域时不得拉伸或扭曲人物;可按比例裁切或自然扩展原有青蓝环境背景。右半将同一位女性、她的姿态与安静直视的情绪重新讲述为稚拙民艺编辑插画。主动删减细节而非完整描摹,使用简化块面剪影、原始化形体、少量色块与手工图符,保留使她一眼可辨的视觉线索。用高低、遮挡、大小和色块关系建立平面叙事,不追求严格透视或解剖;不增加其他人物或虚构场景。采用蜡笔、油画棒、粗铅笔和干性材料般的颗粒肌理。深色区域保留擦拭、叠涂、断裂与纸张透底痕迹;边缘可有手工剪切感、不完美和轻微套印偏差。插画保持平面,不使用写实光影、金属材质、复杂体积或光滑数字矢量。仅从照片中实际可见的颜色提炼简洁明快、克制鲜明的色盘,以暖白纸面和深色形体为基础,适当加入少量提纯的青蓝、浅肤色/奶油色与自然唇色粉红点缀。颜色干净有活力但不过饱和;避免灰脏、沉闷莫兰迪色、荧光色和廉价糖果色。突出人物叙事中心、视觉重心、正负形、大小节奏与呼吸感留白;插画可偏置或局部裁切。只加入少量源自照片可见主体与情绪的文字。中文输出时仅使用一个短情绪标题“静望”和一条极小说明“静谧一瞬”,不虚构地点、事件、日期或人物经历。使用轻巧朴素的民艺编辑字形,带轻微手工套印偏差和不规则字距,让文字自然落在留白或场景结构中,不形成商业标题栏。画面应亲切、幽默、朴素、聪明、有生活感,同时具备成熟设计意识。避免儿童卡通、可爱贴纸、写实插画、复杂透视、精细五官、过度装饰、3D渲染、电商插画感、模板化效果和任何上下布局。

Prompt tiếng Anh

Turn the complete ref-10a-v1 photo into one individual high-end design poster, with no multi-photo collage. Use the complete ref-10a-v1 photo as the sole full-image input and identity reference. Follow the final source correction: use a 7:6 landscape side-by-side layout, exact 1:1 split: left and right each occupy exactly 50% of the frame; never top-bottom. The actual photo shows one adult woman with long center-parted brown hair, a black thin-strap top, direct calm gaze and a deep teal-blue background. On the left, preserve the source photo subject identity, structure, pose, authentic texture, natural light and shadow, and color atmosphere, applying only subtle art-magazine color grading. Fit the photo to the left panel without stretching or distorting the woman; a natural proportional crop or extension of the existing teal environment may be used to fit the panel. On the right, retell the same woman, her posture and quiet direct-gaze mood as a naive folk-art editorial illustration. Subtract detail rather than fully trace the photo; use simplified block silhouettes, primitive forms, a few color blocks and handmade icon-like marks, retaining the visual clues that keep her recognizable. Use flat spatial planes and relationships of height, overlap, size and color rather than strict perspective or anatomy; add no other person or invented scene. Work with crayon, oil-pastel, thick-pencil and dry-material grain. Dark areas should show rubbing, layering, broken marks and paper showing through; edges may be hand-cut, imperfect and slightly misregistered. Keep the illustration flat, without realistic lighting, metallic material, complex volume or smooth digital vectors. Extract a concise, vivid but restrained palette from colors actually visible in the photo: warm white paper and deep dark forms as a base, with a few purified teal-blue, pale skin/cream and small natural lip-pink accents where appropriate. Keep colors clean and lively but not oversaturated; avoid muddy gray, dull Morandi palettes, fluorescent color and cheap candy color. Emphasize the portrait as narrative center, clear visual weight, positive/negative shapes, scale rhythm and breathing whitespace; the illustration may be offset or partly cropped. Include only a small amount of meaningful text derived from the visible subject and mood. For an English-language render, use one short mood-derived title, “Quiet Gaze”, and one very small note, “A study in stillness”; do not invent a location, event, date or biography. Use light, plain folk-art editorial lettering with slight handmade print deviation and irregular spacing, allowing the text to sit naturally in whitespace or along the scene rather than forming a commercial title bar. Keep the result intimate, humorous, simple, clever and lived-in while retaining mature design awareness. Avoid children’s cartoons, cute stickers, realistic illustration, complex perspective, fine facial detail, excessive ornament, 3D rendering, e-commerce illustration style, templates and any top-bottom layout.

Xem mô hình API

Xem thêm prompt hình ảnh và video · Tạo hình ảnh · Tạo video · Tạo trên canvas AI