Have a ComfyUI workflow? Get a readable Workflow Doc in your browser.Make a workflow readable.Try it now ↗
Model repo on Hugging Face Hub.
Runtime profile
Source excerpt
Model repo on Hugging Face Hub.
Estimated VRAM requirement
0 GB estimated
Requirements
2 requirementsAI-generated commentary
AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.
Use this ComfyUI workflow to separate a subject from a video's background: it takes a video, uses RMBG-2.0 to create a foreground mask from one loaded frame, then uses MatAnyone2 for matte and green-screen video outputs and an alpha-transparent WebM output.
The matte and green-screen MP4 outputs are set to save; the alpha-transparent WebM output is set not to save.
Both video loaders currently use `tiktok_huyenmyxinhgai126_7604037870168329480.mp4`; select your clip in both loaders when using your own video.
Suggestion · not verified
Get `workflow_matanyone2_combined_rmbg_2.0.json` from Tyanfarm's Hugging Face repository.
The workflow lists two required node packs: `comfyui-videohelpersuite` and `matanyone`.
`comfyui-videohelpersuite` corresponds to ComfyUI-VideoHelperSuite; check which installed node pack provides `matanyone` before running.
Suggestion · not verified
The RMBG branch is configured to load one frame, while a second video loader supplies frames to MatAnyone2.
Both MP4 outputs use NVENC H.264 and take their frame rate from the loaded video's FPS.
The repository contains the workflow JSON but no model artifact files.
Does this need to be edited?
Sign in to send an edit request.
Sources
1 sourceSource excerpts
1 excerptSource context: 0 downloads · 0 likes · Repo tyanfarm/ComfyUI-workflows
Node pack · Registry
matanyone
PossibleNode pack · Registry
