<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/d950b76c-d10e-4bed-b9d6-05e163d49213/width=525/d950b76c-d10e-4bed-b9d6-05e163d49213.jpeg" />Hey Y'all! Use my brand-new workflow to make your character dance or move to your heart's content using Wan SCAIL! Inputs:<ul><li>Character image</li><li>Video reference</li><li>Resolution (needs to be multiples of 32)</li></ul>Output:<ul><li>Your character moving according to the input video!</li></ul> Adjust your resolution, video length, context window size and block swapping values to maximize quality and length on your system. Currently, its set up for a 24 G VRAM system, 16 G is easy with a slightly smaller resolution. 12 G may be achievable, but not recommended...Notes:<ul><li>I've heard that some people have an extra input for this node. I currently use the SAM3 version 0.1.1. of this custom node, and a latter update may have added an extra input. I will take a look into it.<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/2c08430e-6e33-4c05-b62a-b85615ff47a5/width=525/2c08430e-6e33-4c05-b62a-b85615ff47a5.jpeg" /> </li><li>On the first run, it will fail, but this image will load into the SAM3 Point Collector. Mark out your main character like this:<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/32020c7a-e242-43e5-bfde-efeb26d855ca/width=525/32020c7a-e242-43e5-bfde-efeb26d855ca.jpeg" /></li><li>Only use this if your input image has a solid color background and needs extension. (I will likely remove this in future)<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/02b8039a-3c09-419f-a480-3f17fb367dd6/width=525/02b8039a-3c09-419f-a480-3f17fb367dd6.jpeg" /></li><li>The download of the NFL Model should proceed automatically, if that does not occur, you will to find and manually download it. (WIP)<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/15203eb8-b88a-44e1-807b-bc76a4a36743/width=525/15203eb8-b88a-44e1-807b-bc76a4a36743.jpeg" /></li></ul>Future Versions<ul><li>I'm thinking about removing the padding chunk. Feels extraneous.</li><li>Would be interesting to tinker with Enhance Video, SLG and FreeStar to see if any quality can be squeezed out.</li><li>I will NOT be added BG fusion, nor frame interp, nor upscale in here. Separate workflows are better.</li><li>Would love to explore other methods of improving the facial expressions.... will be difficult without switching models though...</li></ul>V1.0<ul><li>Initial Release! It just works!</li></ul>Enjoy the workflow, happy genning! I found this workflow quite fun and hopefully with other uses other than just dances.Stay Cheesey Y'all!

Brie's Wan Lazy SCAIL

Initial Commit! It just works!

Files with SHA256: b49f3393c6d46ecd6b7071cfa8eba7118d89ec0202c2e9cf94c7fd7c32a06f25