Loading Dashboard...
Preparing your workspace

Genjutsu: Swap Anything in Your Video or Move It Somewhere New

Genjutsu is VdoBloom's video-to-video editor on ByteDance Seedance 2.5. Upload a clip you already have, add one photo per thing you want to change and label it, and the output is the same shot with the change applied — faces, movement, camera and timing exactly as filmed. Object Swap changes what you label and leaves the scene alone; Motion Transfer keeps every person and every move and rebuilds the location from your photos.

Clips are 2-30 seconds at 480p-720p and come back at the same length and ratio in 480p, 720p or 1080p. Up to ten labelled photos per job. The price is quoted from the clip's length the moment you pick the file, and clips outside Seedance's limits are refused before anything is charged.

How it works

  1. 1

    Pick a preset

    Object Swap to change things inside the shot; Motion Transfer to keep the people and action and change the place. Both cards show real outputs from our tests.

  2. 2

    Upload the clip

    MP4 or MOV, 2-30 seconds, 480p-720p. The cost for this exact clip appears immediately.

  3. 3

    Add labelled photos

    One photo per change, each with a label — "the man on the left", "the street". Up to ten.

  4. 4

    Generate

    Four to seven minutes. The result lands in My Creations, watermark-free on paid credits, refunded automatically if the job fails.

Frequently asked questions

What can Genjutsu change in a video?

Anything you can photograph and label: a person, an outfit, a product, a prop, a sign — or, with Motion Transfer, the entire location while the people and their movements stay. Several changes can go in one job.

Does it keep faces and motion?

Yes. Seedance 2.5's editing mode preserves identity, pose, camera path and timing; in our tests a player's kit changed to a leather coat with his face and stride untouched, and a stadium became a container port with all eleven players identical.

How is the price worked out?

Per second of clip on the with-reference-video rate at the output resolution you choose — for example 473 credits for a 15-second clip at 720p. The figure is shown before you confirm and deducted up front.

How is this different from Higgsfield Genjutsu?

Same model family and the same two presets. VdoBloom adds per-photo labels, a price quoted before upload, pre-charge validation and one-time credits that never expire; Higgsfield allows more reference photos per job. The comparison is at /tools/higgsfield-genjutsu-alternative/.