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

砂嵐の竹海で剣を抜く剣客

Swordsman Drawing Blade in a Sandstorm Bamboo Sea

「砂嵐の竹海で剣を抜く剣客」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。

砂嵐の竹海で剣を抜く剣客 — AI image creation
砂嵐の竹海で剣を抜く剣客

プロンプト

中国語プロンプト

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。 创作一幅具有电影感的东亚奇幻插画,描绘一位明确25岁或以上的成年男性;仅使用ref-10a-v1维持可见面部身份,不推断个人经历。他从映照着黎明最后一抹暖光的风化石砌废墟,走向渐隐于冷蓝灰薄雾中的高密竹海,并在两种环境交界处停下,以平静、克制的姿势缓缓拔剑。为他穿上层叠的炭灰色旅行长袍、低调石板灰腰带和实用靴子。让一层克制的风沙掠过废墟一侧,同时确保主体清晰可见,再逐渐融入竹林冷雾;色彩从废墟的暖琥珀晨光过渡到冷雾玉绿与蓝灰色。以高耸竹影和残破石墙围合人物,营造空间纵深与清晰剪影。重点表现静默决意与悬而未决的最后瞬间,而非攻击:剑刃保持受控,不出现攻击对象、伤口、血迹或战斗后果。采用精致的电影感绘画构图,细致表现布料与石材纹理,衣摆和竹叶仅作克制风动。不要未成年人、额外角色、血腥暴力、文字、Logo或水印。

英語プロンプト

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. Create a cinematic East Asian fantasy illustration of one clearly adult man, age 25 or older, using ref-10a-v1 only to maintain visible facial identity without inferring biography. He moves from weathered stone ruins catching the last warm light of dawn toward a dense sea of tall green bamboo fading into cold blue-gray mist, pausing at their threshold in a calm, deliberate sword-drawing pose. Dress him in layered charcoal travel robes with a muted slate sash and practical boots. Let a restrained veil of windblown sand, with clear visibility of the subject, sweep across the ruin side and thin into cool forest mist; grade the scene with warm amber dawn at the ruins shifting into cool misty jade and blue-gray. Frame the figure within towering bamboo and broken stone, with atmospheric depth and clear silhouette. Emphasize quiet resolve and a suspended final moment, not an attack: blade remains controlled and no target, injury, blood or combat aftermath is shown. Refined painterly cinematic composition, nuanced fabric and stone texture, restrained motion in robe hems and bamboo leaves. No minors, extra characters, graphic violence, text, logo or watermark.

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