OpenZhenZhen AI 提示词画廊 · Moss Image · image

成振宇 · 越过屋脊 · 棱镜折射

Sung Jin-Woo · Over the Rooftops · Prism Refraction

Moss Image 原创创意图集｜棱镜折射｜成振宇。原创提示词生成。

成振宇 · 越过屋脊 · 棱镜折射 — AI image creation
成振宇 · 越过屋脊 · 棱镜折射

提示词 / Prompt

中文提示词

创作一张富有观赏性的「棱镜折射」完整照片：成振宇，经典特征：黑色后掠发、亮蓝眼、深长外套与安全低持双弯刃。动作：屈膝越过瓦顶间隙，衣角后扬，落点清楚。场景细节：窗格投影与漂浮尘光。表现形式：精工关节手办的叙事微缩实拍，动作关节自然、场景尺度统一，不露支架。只设一位清楚可辨的经典角色，保留原作标志服装、发型、面容、轮廓及明确点名道具。动作符合该角色真实肢体结构，爪、翅膀或履带不凭空长出人手。武器保持入鞘或安全低持，除非是无害训练动作。摄影：镜头前真实玻璃棱镜形成克制光谱边缘，中心面部保持自然。构图：临场平视中景，可触前景与清楚面部背景分离。环境点缀配色：炭黑与珊瑚红，保留主角经典固有色。捕捉一个具体、有情节、动作可读的关键瞬间，不采用通用站桩姿势。面向网站图库缩略图，设置鲜明唯一主角、清爽轮廓、清楚面容、克制背景细节及有意义的前后景纵深。动作道具与必要肢体完整入画，动作需要时优先放宽景别。重量、接触、双手、衣料运动和反射应可信。电影摄影完成度，独立单幅，不要分格、字幕或水印。

English prompt

Create a visually striking finished prism portrait photography photograph: Sung Jinwoo from Solo Leveling, canonical features: black swept hair, luminous blue eyes, long dark coat, paired curved daggers held safely low. Action: vaulting across a tiled roof gap, knees bent and coat trailing, landing zone visible. Scenic detail: window-grid shadows and suspended dust in the light. Representation: a premium articulated figure photographed in a narrative miniature, natural joint pose, consistent scale, no visible support stand. One principal recognizable canonical character. Retain the original signature costume, hair, face, silhouette and any explicitly named prop. Adapt the action to this character's actual limb structure; do not invent human hands for paws, wings or tracks. A named weapon stays sheathed or safely lowered unless required by a harmless training beat. Photography: a real glass prism near the lens adds a restrained spectral edge, central face remains undistorted. Composition: intimate eye-level medium view, tactile foreground, strong face-to-background separation. Environmental accent palette: charcoal and coral red; retain canonical subject colors. Capture one specific story moment at the peak of a readable gesture, not a generic stock standing pose. For a website gallery thumbnail, establish one strong focal subject, a clean silhouette, clear face, restrained background detail and meaningful foreground-to-background depth. Keep the action prop and essential limbs within the frame; use a wider view if the chosen action requires it. Resolve weight, contact, hands, fabric motion and reflections convincingly. Photographic cinematic finish, one image, no panels, captions or watermark.

主题与风格

探索更多图片和视频提示词 · 生成图片 · 生成视频 · 在 AI 画布中创作