What you'll make

Film yourself once, in any room, and turn it into a vertical video that shows three versions at the same time.

916 Pletor (28).mp4

<aside> 👋

All my tutorials on Youtube

Follow to get the most of workflow’s explanation. Come drop a like & comment !

How to Turn One Video into Multiple Camera Angles with AI (STEP BY STEP GUIDE)

</aside>

You need a phone and a tripod, a Pletor account (it runs Seedance 2.5), a character image and a video editor. Count about an hour for your first one.

My workflow:

Pletor - Character + Scene swap

Step 1: Film one take

Everything later depends on this clip, so spend five minutes getting it right. The AI keeps your gestures, timing and voice, and replaces everything else.

  1. Lock the camera. Phone on a tripod or a stack of books, landscape (16:9). No handheld, no zoom.
  2. Frame from the waist up. Keep your head, both hands and the chair or desk in shot. The AI needs to see your whole body to rebuild a new one on it.
  3. Keep it under 30 seconds. Seedance 2.5 handles up to 30 seconds per generation; under 20 is the easiest.
  4. Record clean audio. A clip-on mic or a quiet room. Your real voice goes on the final video.
  5. Leave tiny pauses between sentences. Those pauses become the camera cuts later.
  6. Plan your transformation word. If you change character mid-video, say a clear word at that moment (for example "become"). You will anchor the swap to it.
  7. Talk with your hands, but keep them away from your face. Hands crossing the face confuse the swap.

The messy room and bad light don't matter. That contrast is part of the reveal.