This is a port to the H3 FL2VA model of <a target="_blank" rel="ugc" href="https://civarchive.com/models/2119841/wan-22-i2v-multi-first-to-last-frame-fflf-video-automated-easy-looping-multi-prompting">Wan 2.2 I2V - Multi First-to-Last-Frame (FFLF) video. Automated, Easy, Looping, Multi-Prompting</a>.It automates the production of any type of long cinematic sequence, such as photoshootings, using the First-to-Last frame technique and a bunch of keyframes.<h2 id="settings">Settings</h2><ul><li>Load Images From Folder (KJ)<ul><li>folder: set the location of your images</li><li>load_image_cap: the number of images to run the workflow on.</li><li>start_index: skip the first n images.</li></ul></li><li>Looping Video<ul><li>if set to true, the workflow will do an extra loop to link the last image of the dataset to the first one.</li></ul></li><li>SetImageSize<ul><li>width and height must be multiples of 32.</li></ul></li><li>Master Prompt<ul><li>Customize the prompt here. It will be injected at every run.</li></ul></li><li>Specific Promps (1 per line)<ul><li>Add prompt instructions for specific runs. The first line will be used at the first run, etc.</li></ul></li><li>Lengths (17n+5)<ul><li>The number of frames to generate. Typically one value of 22, 39 or 56. Can be customized for each run.</li></ul></li></ul><h2 id="usage-tips">Usage Tips</h2><h3 id="ram">RAM</h3>While the workflow can theorically run on an infinite number of images, hardware limitations will hit at some point. For Windows users: make sure to have set up a huge pagefile.sys file before the run.<h3 id="audio">Audio</h3>Since each image pair is generated separately, the audio generated by H3 will be inconsistent. Therefore, the audio outputs are disabled, and replaced by a Load Audio node.<h3 id="transitions">Transitions</h3><ul><li>If they feel unnatural : try changing the order of images, and make sure the various elements, such as the background and the clothing, are consistent from image to image.</li><li>If they feel too slow or too fast : try adjusting the chunks' lengths individually.</li></ul><h3 id="turbo">Turbo</h3>I got the best results using:<ul><li><a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/MiniMax-H3_comfy/blob/main/loras/minimax_h3_fl2v_lightx2v_turbo_4step_v0.1_comfy.safetensors">minimax_h3_fl2v_lightx2v_turbo_4step_v0.1_comfy</a> lora, at 0.75 strength and 8 steps.</li></ul><h3 id="interpolation">Interpolation</h3>It is not essential since the videos are generated at 24 fps, but interpolation helps at smoothing transitions. Feel free delete, or disable and disconnect the subgraph if not used.<h3 id="misc">Misc</h3><ul><li>Skip 1st/last frame: not sure which one works best, but better keeping one of them enabled.</li><li>Save Chunk: records each generated chunk individually.</li><li>Preview Whole : gives a preview of the combinated chunks.</li></ul><h2 id="going-further">Going Further</h2>I believe the generated output, along with the audio and a few keyframes, could be used with the Ref2VA model for further refinements (such as smoothing transitions up and lipsync), but I haven't managed to produce anything meaningful with it to date.<h2 id="dependencies">Dependencies</h2>Required<ul><li><a target="_blank" rel="ugc" href="https://github.com/yolain/ComfyUI-Easy-Use">ComfyUI-Easy-Use</a></li><li><a target="_blank" rel="ugc" href="https://github.com/kijai/ComfyUI-KJNodes">ComfyUI-KJNodes</a></li><li><a target="_blank" rel="ugc" href="https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite">ComfyUI-VideoHelperSuite</a></li></ul>Optional<ul><li><a target="_blank" rel="ugc" href="https://github.com/ClownsharkBatwing/RES4LYF">RES4LYF</a> (to set the output's size)</li><li><a target="_blank" rel="ugc" href="https://github.com/kijai/ComfyUI-GIMM-VFI">ComfyUI-GIMM-VFI</a> (for interpolation)</li></ul>

Files with SHA256: 1066cf87b570a717cac34afbacf3d5eefc68b2552cdb75d4c60ed179fbc4d0ae