Film & CG creation · depth-guided motion transfer

Depth-Guided Motion Transfer

SEELE runs a depth-guided motion-transfer workflow from exactly one reference video and one replacement-subject reference path. It derives depth guidance, then generates the replacement subject while preserving the original performance, camera movement, timing, composition, and audio as closely as the workflow allows.

Add your details here, then continue in Workspace to create.

Create your depth-guided character video

Enter a description of the replacement subject, then continue in Workspace to add exactly one source video and, if needed, a subject image to create a depth-guided motion-transfer video.

What happens next

Generated media may require continuity, rights, factual, and technical review before production use.

Your entries stay in this browser until Workspace opens.

Built for a concrete production decision

Each use case focuses on a practical result you can create in Workspace without promising an unsupported all-in-one pipeline.

VFX concept teams

Test a replacement subject against an existing performance and camera.

Character creators

Carry a supplied subject reference into motion guided by a real video.

Previs editors

Explore character replacement while retaining timing and composition.

From input to a reviewable artifact

Follow these steps to move from your starting material to a result you can review.

Attach one source video

The workflow requires exactly one reference video.

Choose one subject path

Attach one replacement-subject image, or provide a non-empty text description when no image is attached.

Build depth guidance

SEELE converts the source video to a depth-spatial guide for the resulting motion.

Generate and verify

Review the replacement against motion, camera, timing, composition, and audio.

Start with the details that matter

Use this example as a starting point. Add only facts, references, and delivery requirements you can verify.

EXAMPLE
Use my attached source video for motion and depth guidance. Replace the main character with the subject in my attached reference image while preserving performance, camera movement, timing, composition, and audio.
Variant 01
Attached character image replacing the performer
Variant 02
Text-described creature when no image is available
Variant 03
Object replacement preserving camera and timing

See what you can create

These are the results and delivery choices available for this type of project. Confirm exact file formats in Workspace.

Real SEELE Workspace card artwork supporting depth-guided motion transfer — Depth-guided video
Output 01

Depth-guided video

Generated replacement-subject video based on the derived depth guide.

Real SEELE Workspace card artwork supporting depth-guided motion transfer — Depth artifact
Output 02

Depth artifact

The workflow-derived depth.mp4 used as spatial guidance.

Real SEELE Workspace card artwork supporting depth-guided motion transfer — Review checklist
Output 03

Review checklist

Performance, timing, composition, camera, subject, and audio checks.

Questions before you open Workspace

What does this depth-guided motion transfer do?

SEELE runs a depth-guided motion-transfer workflow from exactly one reference video and one replacement-subject reference path. It derives depth guidance, then generates the replacement subject while preserving the original performance, camera movement, timing, composition, and audio as closely as the workflow allows.

How quickly can I start?

Enter the requested details here, then continue in Workspace to create your result. Processing time varies with your request and service conditions.

Can I edit the result?

You can change your entries and create another version. Confirm available editing, export, and integration options in Workspace.

Do I need source files?

Yes. Continue in Workspace to add exactly one source video, plus one replacement-subject image or a written description when no image is used.

What should I check before production use?

Review factual accuracy, rights, identity, continuity, motion, framing, audio, and technical delivery. Generated media can require manual polish.

What still needs judgment

These constraints are part of the product answer, not fine print.

Exact input contractOne reference video is mandatory, plus exactly one image-or-description subject path.
Failure stops the workflowIf reference preparation or depth conversion fails, generation should not continue.
Not perfect captureOcclusion, fast motion, fine limbs, and identity can still need retries or cleanup.

Bring your idea to life in Workspace.

Add any photos, videos, or other references you need, create the result, and review it as you refine.