ワークフローに戻る

Wan2.1 SCAIL

One-Touch SCAIL Pose Control based on the composition of the Reference Image

ワークフローをダウンロードCloudで試す

Input

The character to animate; its framing and look are kept.

Describe the character doing the motion.

A clip of one person moving; only their pose and motion are used.

How many frames of the driving video to follow (4n+1; at 24 fps, 77 is about 3 s). The clip must be at least this long.

Default: 77

Output height (multiple of 32).

Default: 832

Output width (multiple of 32).

Default: 480

Advanced settings

Frames per second of the result.

Default: 24

What to avoid.

Leave empty for a new result each run. Enter a number to repeat one.

Sampling steps (the lightx2v LoRA is tuned for about 6).

Default: 6

Run

Output
00:00
Example

例を試す

このワークフローについて

The Wan2.1 SCAIL workflow in ComfyUI is designed to seamlessly transfer motion and pose information from a video source onto a reference image, while maintaining the original composition of the image. This is achieved using the Wan2.1 SCAIL model, which excels in adapting pose data to fit the proportions and framing of the reference image.

The workflow utilizes several custom nodes, including ComfyUI-WanAnimatePreprocess and ComfyUI-WanVideoWrapper, to preprocess and wrap video data efficiently. These nodes work in conjunction with the comfyui-scail-pose node, which is responsible for the precise scaling and alignment of pose data. By leveraging these tools, users can create dynamic animations that preserve the aesthetic qualities of the original reference image.

This workflow is particularly useful for applications that require high fidelity in pose transfer, such as creating animated sequences from static images or enhancing video content with consistent stylistic elements. The integration of nodes like LoadImage, SaveVideo, and LoadVideo ensures a smooth workflow from input to output, allowing users to easily import their media, apply the SCAIL technique, and export the final animated video. This makes the workflow not only technically robust but also user-friendly, providing a one-touch solution for complex pose control tasks.

よくある質問