With this workflow I was able to create a 1024x1024 12.5s 8fps video in 19min (and 768x768 in 13min) on an RTX 2060 mobile with less than 4GB VRAM.This workflow is intended to create a video morphing between two IPAdapter image models. In addition OpenPose images can be used to support the animation.The workflow iterates through the frames one-by-one with batch size 1 and therefore uses low VRAM.The workflow creates only the png-frames, so the actual video needs to be created with an external tool like ffmpeg:<pre><code>ffmpeg -framerate 8 -pattern_type glob -i 'vid4*.png' vid4.webm</code></pre>Version 2 remarks:<ul><li>the calculation going into the save image filename_prefix should be changed to "a + b.zfill(5)" for easier sorting of filenames.</li><li>Here are some poses used for generating my example video: <a target="_blank" rel="ugc" href="https://civarchive.com/models/329183">https://civarchive.com/models/329183</a> (filenames and paths need to be adjusted)</li></ul>Version 1 remarks:I am using the following OpenPose images in the workflow but it can be adapted to other Poses aswell:<a target="_blank" rel="ugc" href="https://civarchive.com/models/162947/open-pose-dwpose-running-animation-figures">https://civarchive.com/models/162947/open-pose-dwpose-running-animation-figures</a>(path needs to be adapted)The workflow is currently using SVD as CLIP_VISION model, but other CLIP_VISION models/loaders could also be used instead.Version 2 changelog:<ul><li>Improved layout with groups</li><li>normal CLIP_VISION loader instead of SVD</li><li>Poseframe number saved in filename</li><li>Depth controlnet</li><li>Image morphing bypass, Controlnet bypasses</li><li>IPAdapter stength control</li></ul>Version 2.5 (In progress, probably to be released in June):<ul><li>option for more consistency between generated frames</li><li>replacement of hacky jobiterator nodes by either manual queue or loop nodes from impact pack.</li></ul>

ComfyUI image-morphing low-VRAM LCM video animation workflow with poses

<ul><li>Improved layout with groups</li><li>normal CLIP_VISION loader instead of SVD</li><li>Poseframe number saved in filename</li><li>Depth controlnet</li><li>Image morphing bypass, Controlnet bypasses</li><li>IPAdapter stength control</li><li>...</li></ul>

Files with SHA256: c73292cfcd11494b72c385285cfd73955a0596d7da794a7068678364330df9b4