1. Inspiration source
Community example
This tutorial teaches a similar generation technique using VidLuma. Credit public inspiration, replace brand-specific details, and avoid presenting someone else's work as your own original output.
2. Effect breakdown
Output target
Video · 16:9
Use case
Characters
Main tool
Text to Video
Model family
Seedance 2.0, Kling 3.0
3. Prompt recipe
Treat this as a recipe, not a locked script: replace the subject, brand, scene, and ratio while keeping the motion/layout constraints.
10-second nighttime city selfie video, 9:16 vertical, authentic night vlog quality. A clearly adult East Asian woman walking while filming herself with a front-facing phone. The bottom of the frame consistently shows her extended selfie arm. Long hair blown by night wind, plain light knit cardigan and floral dress, fresh and pretty. Setting: nighttime sidewalk under an awning, parked cars, glass reflections, distant building lights forming bokeh, all plates and signs unreadable. Camera: front-facing wide-angle with slight edge distortion, handheld micro-tremor. Timeline: 0-2s tilt up to find her laughing face; 2-5s walks looking at camera, car light trails; 5-7s phone raised, brief eye contact; 7-9s jogs half a step closer, laughing; 9-10s focus lands on her happy expression. Prohibited: text, logos, watermarks, readable plates, bystander faces, excessive beauty filtering, underage appearance, face drift.
4. Recommended settings
ToolText to Video
Aspect ratio16:9 (landscape)
First runGenerate one draft first; create variants only after composition is stable.
Reference ruleUse a clean image reference when possible; avoid overloading camera movement.
5. Step-by-step
1
Seedance 2.0 / Kling 3.0
10s vertical night selfie, handheld micro-tremor, neon bokeh, timeline-controlled shots.
View step prompt
10-second nighttime city selfie video, 9:16 vertical, authentic night vlog quality. A clearly adult East Asian woman walking while filming herself with a front-facing phone. The bottom of the frame consistently shows her extended selfie arm. Long hair blown by night wind, plain light knit cardigan and floral dress, fresh and pretty. Setting: nighttime sidewalk under an awning, parked cars, glass reflections, distant building lights forming bokeh, all plates and signs unreadable. Camera: front-facing wide-angle with slight edge distortion, handheld micro-tremor. Timeline: 0-2s tilt up to find her laughing face; 2-5s walks looking at camera, car light trails; 5-7s phone raised, brief eye contact; 7-9s jogs half a step closer, laughing; 9-10s focus lands on her happy expression. Prohibited: text, logos, watermarks, readable plates, bystander faces, excessive beauty filtering, underage appearance, face drift.
6. Common mistakes
Too broad
“Make it cinematic and cool” gives the model no crop, material, or motion constraints.
Better
Name the subject, ratio, lighting, layout, motion, and what must remain unchanged.
