Do I need to mask the person or use a green screen?

No. The GeminiVideoOmni node uses the source video as structure guidance and aims to preserve the original scene. A clean subject and fewer occlusions will improve results, but no masking is required.

What kind of outfit reference works best?

Use a single, sharp image with the garment clearly visible, minimal wrinkles or motion blur, and a view that matches the pose and angle in your video. Crop away faces and busy backgrounds if possible to reduce ambiguity.

How long and how large can my video be?

Shorter clips (e.g., under 10–15 seconds) are recommended for faster iteration. Longer or high-resolution videos increase compute and time. If you hit memory or time limits, downscale the video or process in segments.

Will it handle loose fabrics, patterns, or accessories?

It can, but complex elements (flowing skirts, transparent fabrics, shiny materials, intricate logos) are harder to track and may flicker. Choosing a clear reference and moderate motion helps improve temporal consistency.

Gemini Omni - Outfit Swap - ComfyUI Workflow

This ComfyUI tutorial walks you through a focused outfit-swap workflow powered by the GeminiVideoOmni node. You load a person video with LoadVideo (Your Footage) and a single outfit photo with LoadImage (Outfit Reference). The GeminiVideoOmni node takes these inputs plus a plain-language instruction (provided via a PrimitiveNode) to map the outfit from the reference image onto the moving subject in your clip, while preserving identity, motion, camera framing, and the original background. The generated result is written out with SaveVideo.

Technically, the workflow conditions a video-aware generative model on both the source frames and the reference outfit image. The source video guides structure and motion so faces, hands, and environment remain stable, while the reference guides the appearance of clothing across frames for temporal consistency. The MarkdownNote nodes in the graph provide inline guidance; they don’t affect processing. With just a few nodes and no manual masking or rotoscoping, this setup is ideal for fast virtual try-ons, fashion concepting, and styling previews.

API

Use this workflow from code

Every Comfy workflow is a JSON graph. The payload below is this workflow, exactly as ComfyUI runs it — fetch it from the URL, keep it in version control, or load it in ComfyUI and run it node by node.

GET https://comfy.org/workflows/download/4600a60d661b.json

Fetching workflow JSON…

Run it from Python or TypeScript with the Comfy SDK. The same code targets Comfy Cloud or a ComfyUI you host yourself — only the base URL changes.

# Install (beta)
pip install comfy-sdk        # Python
npm i @comfyorg/sdk          # TypeScript

# Run this workflow (Python)
from comfy_sdk import Comfy

client = Comfy(api_key="comfyui-...")
wf = client.workflows.from_file("workflow_api.json")
job = client.run(wf)
for output in job.get_outputs("<output-node-id>"):
    output.to_file(output.name)

The SDK takes a workflow in API format: open this workflow in ComfyUI and use File → Export Workflow (API). API access requires a Comfy Cloud plan with an API key. SDK docs · Get an API key

FAQ

Frequently Asked Questions

View all workflows