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

オリジナル白金ショートヘアSF剣士の廃墟都市訓練肖像

Original Platinum Short-Haired Sci-Fi Swordswoman Training Portrait in Ruined City

「オリジナル白金ショートヘアSF剣士の廃墟都市訓練肖像」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。

オリジナル白金ショートヘアSF剣士の廃墟都市訓練肖像 — AI image creation
オリジナル白金ショートヘアSF剣士の廃墟都市訓練肖像

プロンプト

中国語プロンプト

使用参考人物 ref-10a-v1 作为原创角色的面部参考,创作一张2:3竖版写实科幻动作肖像。主体为明确成年的原创女性剑士,白金色下巴长度短发带柔和蓬松度,侧分刘海由细黑发带固定,冰蓝色眼睛保持温柔但专注的训练表情,不对应任何现成角色或真人cosplayer。她保持有力但安全的宽步训练姿势,膝盖微弯、重心稳定,双手戴黑色手套高举一把无锋训练剑,准备完成受控的向下练习动作,不造成伤害。她穿完全不透明的黑色高领战术外套、结构化长裤、白色护臂和黑色防滑战术靴,服装端庄、功能性强、非性化,不使用开口、露肤或透明材质。场景是荒凉的废墟城市,裂纹泥地、碎石、扭曲钢筋和小水洼位于前景,后方是露出钢筋的混凝土塔楼与沉重灰色雷云。左前方漫反射阴天光形成柔和阴影,刀刃边缘和肩部有受控冷色高光,冷青轮廓光勾勒人物。使用85mm f/2.8视角、低但舒适的膝高机位、居中垂直构图和浅景深,以板岩灰、木炭黑、纯白和青色阴影为主的去饱和电影色彩。避免现成角色、cosplayer复制、名人、血腥、伤口、致命伤害、透视服装、内衣、品牌、文字、标志、水印和解剖错误。

英語プロンプト

Use reference person ref-10a-v1 as a face reference for an original character and create a realistic 2:3 vertical sci-fi action portrait. The subject is a clearly adult original female swordswoman with a platinum chin-length bob, soft volume, side-swept bangs held by a thin black headband, and ice-blue eyes with a gentle but focused training expression. Do not depict any existing character or real cosplayer. She holds a strong but safe wide-footed training stance with slightly bent knees and stable weight, raising a blunt practice sword overhead with both gloved hands for a controlled downward practice motion, without causing harm. Dress her in a fully opaque black high-neck tactical jacket, structured trousers, white arm guards, and black non-slip tactical boots; the styling is functional, modest, and non-sexualized, with no cutouts, exposed body, or sheer material. Set the scene in a desolate ruined city with cracked earth, rubble, twisted rebar, and small puddles in the foreground, and concrete towers with exposed rebar beneath heavy gray storm clouds behind her. Use diffused overcast light from the front left, soft shadows, controlled cool highlights on the blade edge and shoulders, and a cool rim light. Use an 85mm f/2.8 perspective, a low but comfortable knee-height camera, centered vertical composition, and shallow depth of field. Grade in desaturated slate gray, charcoal, stark white, and teal-shadow tones. Avoid existing characters, cosplayer replication, celebrities, gore, wounds, lethal harm, sheer clothing, lingerie, brands, text, logos, watermarks, and anatomy errors.

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