Innovate Futures @ Benji
patreon
Stable Diffusion Animation Create Tiktok Dance AI Video (Tutorial Guide)
🕑 Added 2023-12-06 16:49:24 +0000 UTC
Video : https://youtu.be/wFahkr-b7HI
In today's tutorial, I'm pulling back the curtains on how I create those mesmerizing Tiktok dance videos using incredible Custom Nodes of Stable Diffusions ComfyUI. ComfyUI Flicker-Free + LCM Animation Workflow : https://www.patreon.com/posts/94197439 ComfyUI Flicker-Free Animation Workflow : https://youtu.be/LDeFP6d-4YA Install LCM Lora Full Guide : https://youtu.be/y0Tw9Zb4Sy4 Previous Automatic 1111 Animation Dance Video : https://youtu.be/lrFcvq-BI84 Full Tutorial Of Using Mov2Mov : https://youtu.be/wnlFV9uE-Gg Hey everyone, in today's video, we're taking the ComfyUI AnimateDiff flickering-free workflow to the next level by adding new custom nodes. We'll be using a boxing action stock video footage as our source and exploring two methods to enhance the animation quality. First, we'll discuss using an Upscaler to improve image quality. By connecting the VAE decode output to a high-quality Upscaler model, we can enhance each frame of the animation. This method not only upscales the video size but also improves color and sharpness. Next, we'll dive into the second method, which involves integrating the IP Adapter and adjusting the sampling settings. By tweaking the sampling step and denoising values, we can achieve even better results. We'll explore different settings and models to create unique and captivating animations. So, whether you're looking to enhance image quality with an upscaler or experiment with sampling and denoising for better results, this video is packed with valuable insights. Make sure to subscribe to our channel and leave a comment if you have any questions or suggestions. Let's get started and take your AnimateDiff workflow to new heights!
Comments
Benjamin Law
don't use Pinoklo , directly install the latest version of ComfyUI
Phil Digital
awesome video. can you help me with following error? its apears after controlnet finish its job and i see its result in preview, and i think its from ipadapter and i tried everything but cant fix it ERROR:root:!!! Exception during processing !!! ERROR:root:Traceback (most recent call last): File "D:\PINOKIO\api\comfyui.pinokio.git\ComfyUI\execution.py", line 153, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) File "D:\PINOKIO\api\comfyui.pinokio.git\ComfyUI\execution.py", line 83, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True) File "D:\PINOKIO\api\comfyui.pinokio.git\ComfyUI\execution.py", line 76, in map_node_over_list results.append(getattr(obj, func)(**slice_dict(input_data_all, i))) File "D:\PINOKIO\api\comfyui.pinokio.git\ComfyUI\custom_nodes\ComfyUI_IPAdapter_plus\IPAdapterPlus.py", line 463, in apply_ipadapter self.ipadapter = IPAdapter( File "D:\PINOKIO\api\comfyui.pinokio.git\ComfyUI\custom_nodes\ComfyUI_IPAdapter_plus\IPAdapterPlus.py", line 176, in __init__ self.image_proj_model.load_state_dict(ipadapter_model["image_proj"]) File "D:\PINOKIO\api\comfyui.pinokio.git\ComfyUI\env\lib\site-packages\torch\nn\modules\module.py", line 2152, in load_state_dict raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format( RuntimeError: Error(s) in loading state_dict for Resampler: size mismatch for proj_in.weight: copying a param with shape torch.Size([768, 1280]) from checkpoint, the shape in current model is torch.Size([768, 1024]).
Benjamin Law
1- If appearing as a red box, means your ComfyUI are missing that Custom Node install. You can click the Comfyui manager , click install missing node. Then it will show a list of missing Custom Nodes you need to install. 2 - the resolutions depends on your source video, landscape or portrait, but for faster processing, I will keep 512 , 960 , or 1024 px. Then you can upscale it when you have more time.
James Unsworth
I'm having a few issues running the workflow. I'm using comfy UI with run diffusion. Ttn node pre_text node is appearing as a red box. Also is there a specific input resolutions to use?
Benjamin Law
Hahhaa 😂 I know you guys like this.
Dungeon Master
Excited to watch!
