1 / 2

视频

0 53

Tokyo Skyline Rollercoaster POV

C

作者

C Dance AI

提示词

VIDEO · 1

{ "location": "Tokyo Cityscape (Night)", "duration": "10s", "prompt": "A cinematic POV shot riding an invisible rollercoaster through Tokyo at night. A glowing neon rail 'creates itself' milliseconds before the camera hits it, weaving through the steel structures of Tokyo Tower and nearby buildings. As the camera passes, each building it touches instantly transforms into a stack of glowing cubes that rotate and re-assemble. The shot ends with the camera diving straight down into a sea of neon lights, which turns into a giant QR code or a logo just before the screen goes black.", "vfx_focus": [ "Procedural rail generation", "Dynamic environment transformation (Geometry nodes style)", "Extremely high-speed camera motion with light streaks" ] }

提示词结构分析

这条视频提示词约 773 个字符，由 14 个描述段组成。它围绕“Tokyo Skyline Rollercoaster POV”组织画面，并明确覆盖了运镜、光线、动作、视觉风格、场景。

已包含的控制维度

运镜光线动作视觉风格场景

如何改成你的版本

先替换“Tokyo Skyline Rollercoaster POV”中的主体或场景，同时保留原有动作逻辑。
保留运镜描述，只调整速度、方向或景别，避免一次改变太多变量。
根据成片目标补充声音，并删除不会改变画面的形容词。

更多灵感

Epic anime battle scene, fierce young man warrior with b...

Ultra wide cinematic aerial shot during blue hour over th...

Cinematic Gaming Short with Seamless Transitions

Masterpiece, best quality, official anime character desig...

Main subject: young American woman, early 20s, natural ev...

Stylized 3D animation with exaggerated cartoon proportion...

看得见生成效果的 Seedance 2.0 提示词案例

每个案例都同时提供完整提示词和可播放视频，方便在生成前判断运镜、节奏、光线、主体一致性与视觉风格。

从提示词灵感到 AI 视频的一体化流程

选择提示词后可继续修改场景，按需上传参考图，并在右侧生成器中设置比例、分辨率、时长与音频。

如何写出更好的 Seedance 视频提示词

依次说明主体、动作、环境、镜头运动、景别、光线、视觉风格和时间节奏，保留真正会改变画面的具体信息。