Galerie de prompts IA OpenZhenZhen · gpt-image-2.5 · image

Portrait de dame en hanfu, pinceau fin et couleurs riches, format ultra-vertical 1:3

1:3 Ultra-Vertical Fine-Brush Heavy-Color Hanfu Lady Portrait

Portrait de dame en hanfu, pinceau fin et couleurs riches, format ultra-vertical 1:3 — découvrez cette création avec gpt-image-2.5 sur OpenZhenZhen, son prompt original et un exemple de génération à réutiliser.

Portrait de dame en hanfu, pinceau fin et couleurs riches, format ultra-vertical 1:3 — AI image creation
Portrait de dame en hanfu, pinceau fin et couleurs riches, format ultra-vertical 1:3

Prompt

Prompt chinois

以用户上传的参考图作为主要可见人物的面部身份依据。在不与身份参考冲突的情况下，保留来源的场景、造型、服装、姿势、构图、光线和视觉风格。 将人物渲染为风格化插画，只保留参考图的面部身份特征；不要生成写实摄影人像。 主体：[中国古典女性 / 角色] 花卉：[梅花 / 玉兰 / 紫藤 / 桂花] 主色调：[深靛蓝 + 绯红 / 翡翠绿 + 象牙白 / 暗紫 + 薰衣草紫] 服装：[汉服颜色与面料] 氛围：[宁静 / 优雅 / 梦幻 / 克制] 纵横比：1:3 超纵向 创作一张高度精细的中国风肖像，采用优雅的 1:3 超纵向格式，视觉感受更像是一张长形插画书签，而非标准海报。 保持角色为明确的视觉焦点，占据图像高度的约 75–80%。发型、手部、飘逸的袖子和主裙摆应完整保留在画面内。人物应显得突出，但仍为花卉和背景留出足够的呼吸空间。 垂直构建构图：花枝或垂下的花朵构筑上方和侧边区域，脸部成为中上部分的视觉焦点，长袖、垂坠的面料和裙摆将视觉流动感向下延伸。 将花卉作为构图的一部分，而不仅仅是装饰。创造清晰的前景、中景和背景层，一些枝条或花瓣自然地交错在人物前方，但不遮挡脸部。 为角色穿着层次感丰富的飘逸汉服，带有优雅的长袖和精致的面料纹理。使用精细、连续的金线勾勒服装的关键边缘、主要褶皱、袖子曲线和裙摆的下垂感。金线应遵循织物的自然结构，绝不能变成厚重的边框或密集的全身装饰。 结合精致的工笔风人物表现与浓郁的矿物颜料色块、岩彩质感、细微的调色刀厚涂效果和克制的零星金粉。保持面部、手部、头发和服装更整洁细腻，同时允许背景带有更强的纹理。 使用深邃、饱和的背景颜色使浅色的人物脱颖而出。仅添加少量暗哑的金箔、发光微粒和精细的金色点缀，以连接花卉、配饰和衣物。 最终图像应具有优雅、绘画感、奢华和手工感，带有强烈的垂直节奏，其细节程度足以作为独立艺术品、插画书签、社论视觉图或文化品牌形象。 无文字，无标题，无标志，无编号，无边框，无产品模型，无可见的书签孔或流苏，无现代海报布局，无 CGI 效果，无肮脏磨损纹理。

Prompt anglais

Use the uploaded reference image as the facial identity reference for the main visible person. Keep the source scene, styling, clothing, pose, composition, lighting and visual style unless they conflict with the identity reference. Render the person as a stylized illustration that preserves the reference facial identity cues; do not turn the result into a photorealistic camera portrait. SUBJECT: [classical Chinese woman / character] FLOWER: [plum blossom / magnolia / wisteria / osmanthus] MAIN COLORS: [deep indigo + crimson / jade teal + ivory / dusky violet + lavender] OUTFIT: [hanfu colors and fabric] MOOD: [serene / elegant / dreamy / restrained] ASPECT RATIO: 1:3 ultra-vertical Create a highly detailed Chinese-inspired portrait in an elegant 1:3 ultra-vertical format, with the visual feeling of a long illustrated bookmark rather than a standard poster. Keep the character as the clear focal point, occupying around 75–80% of the image height. The hairstyle, hands, flowing sleeves and main skirt should remain fully visible inside the frame. The figure should feel prominent but still leave enough breathing room for flowers and background. Build the composition vertically: floral branches or hanging blossoms frame the upper and side areas, the face becomes the main focal point in the upper-middle section, and the long sleeves, draped fabric and skirt continue the visual flow downward. Use flowers as part of the composition rather than simple decoration. Create clear foreground, middle-ground and background layers, with a few branches or petals crossing naturally in front of the figure without covering the face. Dress the character in layered flowing hanfu with elegant long sleeves and refined fabric textures. Use fine, continuous gold linework to trace key garment edges, major folds, sleeve curves and the downward flow of the skirt. The gold lines should follow the natural structure of the fabric, never becoming thick borders or dense all-over decoration. Combine refined gongbi-inspired figure rendering with rich mineral-pigment color fields, rock-painting texture, subtle palette-knife impasto and restrained scattered gold dust. Keep the face, hands, hair and clothing cleaner and more delicate, while allowing the background to carry stronger texture. Use deep, saturated background colors to make the lighter figure stand out. Add only small amounts of muted gold leaf, luminous particles and fine gold accents to connect the flowers, accessories and clothing. The final image should feel elegant, painterly, luxurious and handcrafted, with strong vertical rhythm and enough detail to work as a standalone artwork, illustrated bookmark, editorial visual or cultural-brand image. No text, no title, no logo, no numbering, no border, no product mockup, no visible bookmark hole or tassel, no modern poster layout, no CGI look, no dirty distressed texture.

Voir le modèle API

Découvrir plus de prompts image et vidéo · Générer des images · Générer des vidéos · Créer sur le canevas IA