[BONUS TUTORIAL] Create mind-blowing AI RENDERINGS of your 3D animations!

Austen Cope

i am using the SIMPLE video

Austen Cope

I keep getting the mask and the passes i gave it as a result

CZARIO

I forgot to specify I use the model : Mickmumpitz_AI-RENDERING_SDXL_ADV_v07

CZARIO

Hello, I downloaded your workflow and I created a scene in blender, however I cannot get the mask extraction to display in the extract mask section. I have the same color parameters on my objects as in your video but nothing is displayed in the mask preview. Can you help me? Thank you.

Jeff h

Hi, I'm getting the following: ControlNetLoaderAdvanced 'ControlLora' object has no attribute 'device' for both the Canny and Depth ControlNet models. I've tried other canny and depth models that seem to work; however, the prompts I input doesn't really change the images much. I've also tried to add my own LoRA model which was built with Flux Dev, however, it doesn't seem to get the item I built the LoRA for into the image.

Jean-Philippe Rivière

Ok thank you ! Very nice approach, very inspiring !

Jean-Philippe Rivière

Solution : display all custom node in use in the node manager and deactivate them, one of them is breaking the default load image

Mick Mahler

Can you please download the newest version of the workflow? This error is fixed there!

Jean-Philippe Rivière

Same for me, the exact message is : Warning: Missing Node Types When loading the graph, the following node types were not found: Integer... it is a node type...

Matthew

Thank you! Yeah that got it working. Weird that such a basic node would be errored out. But the workflow works now :)

Yonni de Haar

In case you're still struggling with this, just add a 'primitve node', right click on the 'upscale image' node and hit collapse. Then connect one primitive to the width and the other to the height. This automatically makes it an integer.

Matthew

Thank you for this. I have installed everything but when I load your workflow, the "Integer" custom node is missing. What am I missing?

Benlloyd Goldstein

any idea what this error means ? Error occurred when executing ControlNetLoaderAdvanced: Error while deserializing header: HeaderTooLarge File "C:\ComfyUI\ComfyUI_windows_portable\ComfyUI\execution.py", line 151, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\ComfyUI\ComfyUI_windows_portable\ComfyUI\execution.py", line 81, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\ComfyUI\ComfyUI_windows_portable\ComfyUI\execution.py", line 74, in map_node_over_list results.append(getattr(obj, func)(**slice_dict(input_data_all, i))) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Advanced-ControlNet\adv_control\nodes.py", line 91, in load_controlnet controlnet = load_controlnet(controlnet_path, timestep_keyframe) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\ComfyUI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Advanced-ControlNet\adv_control\control.py", line 460, in load_controlnet controlnet_data = comfy.utils.load_torch_file(ckpt_path, safe_load=True) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\ComfyUI\ComfyUI_windows_portable\ComfyUI\comfy\utils.py", line 14, in load_torch_file sd = safetensors.torch.load_file(ckpt, device=device.type) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\ComfyUI\ComfyUI_windows_portable\python_embeded\Lib\site-packages\safetensors\torch.py", line 308, in load_file with safe_open(filename, framework="pt", device=device) as f: ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Siu To Victor Fung

I would like to why there are lots of noise in the mask preview,

The Artist

Seems Like I found a fix, You need to keep the rendered frames in a proper no., for example - If you have an animation of 120 frames, Then Keep 60 instead of 10 for the image load cap. I had 137 frames as my base and tried using less no. of frames (It didn't work, so i removed 17 frames and used 120 Frames with 60 Image load cap.

mortas may

Hello - how do you do basic composting in comfy ui , I have an image with a alpha rendered out in blender , I load image , but what node do I use for the alpha ? Seems so basic but I can’t find anything online , 🤔 Just need simple A/ B then over C ( background)

George Tomlin

Yeah whenever I change the image load cap above 10, I get this error everytime. Any idea what I am doing wrong?

George Tomlin

Hello, I keep getting this error? Any ideas what I am doing wrong? Error occurred when executing KSamplerAdvanced: The size of tensor a (8) must match the size of tensor b (0) at non-singleton dimension 0

John Happé

Thanks a lot for the grat work you are doing !

Mick Mahler

Thank you! :) You can find it under the link in the post. It's called "Mickmumpitz_AI-RENDERING_SDXL_ADV_v02".

George Tomlin