PROMPT RADAR / VIDEO视频叙事
个人资料 / 头像 - Modern Anime Fashion Character
Creates a clean 16:9 anime fashion portrait from a loose Japanese graphic reference, useful for music-video character visuals or concept art.
@HarutaGenAI 0 热度 0 浏览
ORIGINAL VISUAL / 来源见文末编辑先看
它为什么值得进入灵感库
它最有价值的部分，是把一个模糊想法推进到了可以被模型执行的程度。 把一段描述拆成起始状态、动作变化、镜头响应和结束画面，连续性比形容词更重要。
01 / 原始方法
动作与镜头 Prompt
Using REFERENCE_0 only as a loose mood reference for a Japanese retro/graphic sensibility, create a clean anime-style character shot instead of typography. Replace all lettering with one original young woman centered on a plain light gray studio background. Character: a serious, fashionable anime woman with {argument name="hair color" default="black"} shoulder-length bob hair and straight bangs, facing forward with a calm neutral expression. Wardrobe/accessories: include exactly 4 visible fashion elements — 1 oversized black open jacket, 1 fitted white ribbed tank top, 1 pair of high-waisted black pleated trousers, and 1 pair of silver hoop earrings. Composition: medium full-body crop from head to upper thighs, centered slightly high, hands tucked in pockets, minimal negative space, cinematic 16:9 frame. Style: polished modern anime illustration with crisp line art, soft cel shading, subtle highlights on black fabric, natural skin tones, no text, no logo, no watermark.
02 / 复刻路径
不要一次改完所有变量
用 Seedance 复刻时，建议先单独生成稳定首帧，再把动作拆成“开始、变化、收束”三段；每段只保留一个主要运动。
- 主体：Creates a clean 16:9 anime fashion portrait from a loose Japanese graphic reference, useful for music-video character visuals or concept art.
- 视觉方向：社区实测摄影提示词；按原提示词的光线描述执行
- 镜头与构图：按原提示词的镜头与景别执行；按原提示词的构图与画幅执行
- 交付约束：auto，遵循原提示词中的避免项；不要水印、乱码、低清、畸形手、错误透视或主体漂移。
负向约束
遵循原提示词中的避免项；不要水印、乱码、低清、畸形手、错误透视或主体漂移。
模型建议
Seedance / Veo / Kling。先用原始 Prompt 建立基准结果，再根据模型对自然语言、镜头运动或文字排版的能力做局部改写。
来源与编辑原则