OpenZhenZhen AI プロンプトギャラリー · gpt-image-2.5 · image

女性漢服図鑑ポスター

Illustrated Guide to Women's Hanfu Poster

「女性漢服図鑑ポスター」は gpt-image-2.5 を使用した作品です。OpenZhenZhen で元のプロンプトと生成例を確認し、創作の参考にできます。

女性漢服図鑑ポスター — AI image creation
女性漢服図鑑ポスター

プロンプト

中国語プロンプト

一次生成4张彼此独立、完整可传播的4:3横版高端海报,属于同一系列,不做拼贴、四宫格或一张图内的四套服装。若用户未修改{dynasty},默认使用明代;根据该朝代自动选择4种具有代表性、结构差异明显且符合历史服饰逻辑的女子汉服制式,不能只换颜色。每张海报中央放置一位端庄优雅的中国成年女性,完整穿着该张对应制式;将发饰、里衣/中衣、主体上装、下装、外搭、腰部结构、配饰和鞋履等层次,按照该制式真实关系,以精密、柔性、克制的悬浮爆炸图围绕人物轻微分离展开。保持布料柔软垂坠与东方气韵,同时具有产品级结构秩序,四张不能套用完全相同的拆解逻辑。背景使用柔白、米白或浅暖灰白,大面积留白,配合极细引导线、圆点、简短标注、整齐排版、干净通透光线和苹果发布会/高端品牌信息图般的高级传播感,但不要复制任何品牌标志。系列统一字体层级、间距、背景、光影、标注语言和标题系统,同时在制式结构、层次、姿态、发型、低饱和主色和关键词上清晰区分。每张顶部必须渲染准确可读的标题格式“汉服图鉴|{制式名称}”,并配短副标题如“层层有序”或“结构之美”;底部或下方设置简洁信息区,标明朝代、制式、3至5个核心特征和一句文化总结,文字少而清晰,不虚构品牌标记。默认明代可从明制袄裙、马面裙搭配、比甲搭配、正式披风搭配中选择,但当{dynasty}改变时应自动改选符合该朝代的制式。真实呈现丝绸、纱、织物、刺绣、褶皱、系带、玉饰和金属配件等高端材质细节;避免廉价戏服、错误朝代混搭、二次元、网红脸、影楼古风、厚重古籍装饰、教材式满版文字、拥挤长段、杂乱漂浮层次、重复服装、水印、不可读文字、伪品牌标识、人体变形或缺少服装结构。人物必须成年、姿态自然安静、端庄含蓄。

英語プロンプト

Create four independent finished 4:3 landscape posters in one coherent series, not a collage or four-panel composite. Use the default dynasty Ming dynasty unless the user changes {dynasty}; automatically select four representative, visibly different women's Hanfu systems appropriate to that dynasty, never merely recoloring one outfit. Each poster centers one elegant adult Chinese woman in a complete outfit, with her clothing layers gently separated around her in a precise, floating exploded-view arrangement: hair and ornaments, inner garment, main upper garment, lower garment, outer layer, waist structure, accessories, and shoes, adapted to the actual outfit rather than copied across all four. Preserve soft fabric drape and Eastern elegance while maintaining product-grade structural order. Use a bright warm-white, ivory, or pale warm-gray background with generous whitespace, fine guide lines, dots, concise labels, restrained alignment, clean lighting, and premium Apple-keynote / high-end brand information-graphic clarity without copying a logo. The four posters share typography, spacing, background, light, annotation language, and title system but clearly differ in garment construction, layers, pose, hairstyle, accent palette, and feature keywords. At the top of every poster render the exact readable title format '汉服图鉴|{制式名称}' plus a short subtitle such as '层层有序' or '结构之美'. Add a small information area for the dynasty, outfit name, three to five key features, and one concise cultural summary; keep text sparse and legible, with no invented brand marks. Default Ming choices may include Ming aoqun, mamianqun ensemble, bijia ensemble, and formal pifeng ensemble, but select historically coherent alternatives when the dynasty changes. Use realistic high-end textile detail—silk, gauze, woven cloth, embroidery, pleats, ties, jade and metal accents—without cheap costume styling, anachronistic mixing, anime, influencer face, studio glamour, or heavy antique decoration. The woman is calm, poised, modestly elegant, naturally posed and fully adult. Avoid cluttered textbook pages, dense paragraphs, chaotic floating layers, repeated outfits, wrong-dynasty combinations, watermark, illegible text, logo-like marks, distorted anatomy, or missing clothing structure.

API モデルを見る

画像・動画プロンプトをもっと見る · 画像を生成 · 動画を生成 · AI キャンバスで作成