Animate portraits with an input video and a reference image using X-Portrait in ComfyUI.
- source_image: the reference image for generation, should be square and max 512x512.
- driving_video: the driving video containing a face, should match the resolution of the source_image and max 512x512.
- seed: controls current generation seed
- ddim_steps: how many samples (steps) for generating the output. Diminishing returns after 15.
- best_frame: which frame of the driving_video most closely matches the source_image.
I created a custom ComfyUI environment for testing out X-Portrait nodes:
akatzai/comfy-env-xportrait:latest
Create a new environment and copy and paste the link above into the "Custom Image" field in my Environment Manager tool: https://github.com/akatz-ai/ComfyUI-Environment-Manager
Make sure to select the Basic environment type to access the included workflow!
- ComfyUI Manager:
- This node pack is available to install via the ComfyUI Manager. You can find it in the Custom Nodes section by searching for "X-Portrait" and clicking on the entry called "X-Portrait Nodes".
- Clone the repository:
- Navigate to ComfyUI/custom_nodes folder in terminal or command prompt.
- Clone the repo using the following command:
git clone https://github.com/akatz-ai/ComfyUI-X-Portrait-Nodes.git- Restart ComfyUI