分镜预览四格分镜展示提示词中的画面方向,实际视频效果以生成结果为准。
- 模型
- sora-2
- 时长
- 4s
- 画幅
- 21:9
- 分辨率
- 720p
- 参考图片
- 0
- 参考音视频
- — / —
怎样让镜头更好Sora 2 的「制作蓝皮书」级提示词模板:Format → Lenses → Grade → Lighting → Location → Wardrobe → Sound
生成后检查什么本站尚未实测。检查主体一致性、运镜连续性与动作完整性;参考原始来源核对效果。
社区来源 · 本站未实测OpenAI Cookbook ↗ · cclank/lanshu-awesome-ai-video-kit
MIT (repository)
完整提示词 v1
Format & Look: Duration 4s; 180° shutter; digital capture emulating 65mm photochemical contrast; fine grain; subtle halation on speculars; no gate weave.
Lenses & Filtration: 32mm / 50mm spherical primes; Black Pro-Mist 1/4; slight CPL rotation to manage glass reflections on train windows.
Grade/Palette: Highlights—clean morning sunlight with amber lift. Mids—balanced neutrals with slight teal cast in shadows. Blacks—soft, neutral with mild lift.
Lighting & Atmosphere: Natural sunlight from camera left (07:30 AM). 4×4 ultrabounce silver; negative fill from opposite wall. Gentle mist; train exhaust drift.
Location & Framing: Urban commuter platform at dawn. Foreground—yellow safety line, coffee cup. Midground—silhouetted passengers in haze. Background—arriving train braking.
Wardrobe: Mid-30s traveler, navy coat, backpack, holding phone loosely.
Sound: Diegetic only—rail screech, train brakes, muffled announcement, ambient hum, footsteps, paper rustle.
来源模型:Sora · sora-2
原始参数:{"aspect": "21:9", "duration": "4s"}
示例分类不代表本站已接入该模型。实际可用模型以视频片场为准;跨模型使用提示词时,请检查参数并调整。
复用会带入提示词、模型、时长、画幅和已发布的参考素材。缺少的参考文件可在片场补充;提交前会显示费用。
阅读视频创作指南