Cinematic Personal Vlog Scene Generation

A video generation prompt for Seedance 2.5 designed to create a 15-second vlog-style video featuring a specific character reference, simulating natural smartphone filming with handheld shakes.

Cinematic Personal Vlog Scene Generation — poster framePoster frame — the clip lives on the author's original post
Prompt by @eijo(SAT577)View the original post ↗Published 2026-09-06CC BY 4.0

The prompt

Amateur footage shot by a friend on a smartphone. Natural camera shake as if shot handheld, amateur composition, slight focus deviations, natural conversation and interaction, and an emotional atmosphere are created.

<Person1> corresponds to @Image1. Hereafter in this prompt, only <Person1> will be written without @, and all parts written as <Person1> refer to this referenced person.

The 15-second video is a Vlog video capturing one day in the life of the protagonist <Person1>.
Using the following information about the woman, create a Vlog video that summarizes what happened that day like a daily diary in multiple scenes.
The point is to make content full of the woman's charm that will make her fans even more devoted. The costume for <Person1> will be optimized for each scene based on the persona information of <Person1>.

-- Information for <Person1> --
Appearance of the woman:
Refer to <Person1>. Take only the face, hairstyle, skin, and body type from the reference. Do not use the clothing in the reference image.
Maintain consistent facial and physical appearance for <Person1> throughout the entire film.

Persona information:
<Persona information prepared by yourself>
-----------------------

Prohibited items:
Existing anime and products are prohibited. Do not use subtitles. Existing logos, captions, etc., are prohibited.

Copy this into the video model you use — Pixel Shine does not generate video yet.

Model
Seedance 2.0
Media
Video
Language
Japanese
Words
205
Vlog / Social LifestyleAnimeCinematic RealisticPerson / Character

How to use this prompt

  1. 1Copy the prompt — fill in any placeholders first and the copy takes your version.
  2. 2Paste it into the video model you use. Timeline prompts work best pasted whole, not split up.
  3. 3Generate, then judge the first and last second: that is where drift shows up.
  4. 4Bring the result back here to upscale a frame, cut a thumbnail, or clean up a still.

Go beyond generating video

Everything after the generation, on the same site.

Upscale to 4K →Remove the background →Crop for a platform →Compress before sharing →All AI apps →

More from Seedance 2.0

Same model, same dataset — Seedance 2.0 prompts people have shared.

View all Seedance 2.0 prompts →

All video categories

Use cases
Styles
Subjects