Error occurred when executing SEGSDetailerForAnimateDiff: can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first. File "G:\Blender_ComfyUI\ComfyUI\execution.py", line 154, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) File "G:\Blender_ComfyUI\ComfyUI\execution.py", line 84, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True) File "G:\Blender_ComfyUI\ComfyUI\execution.py", line 77, in map_node_over_list results.append(getattr(obj, func)(**slice_dict(input_data_all, i))) File "G:\Blender_ComfyUI\ComfyUI\custom_nodes\ComfyUI-Impact-Pack\modules\impact\segs_nodes.py", line 204, in doit segs = SEGSDetailerForAnimateDiff.do_detail(image_frames, segs, guide_size, guide_size_for, max_size, seed, steps, cfg, sampler_name, File "G:\Blender_ComfyUI\ComfyUI\custom_nodes\ComfyUI-Impact-Pack\modules\impact\segs_nodes.py", line 183, in do_detail cropped_image_frames = cropped_image_frames.numpy() 😂😂😂😂
Are you able to get batch prompts/ list of prompts to work normally with text to image in comfyui like with automatic1111, I know there's the custom nodes for it but it doesn't work , you may need the only man I can ask ( que batch prompt next to que prompt is the closest kind of solution
I used ComfyUI_FizzNodes batch prompt schedule (1:19) here for inpainting, you can see both inpainting and face detailer get it, and the blinking looks good, though I haven't tried text to image yet. Check FizzNodes github, they have bunch of examples and advanced tips for customization.
How can i actually use "face detailer" node with animateddiff? It says it requires 8 latents but when i convert the image batch to image list which face detailer only accepts image lists or single images it changes it to 1 latent and causes the face to look like noise from the lacking latents.
use Simple Detector for AnimateDiff (SEGS), Detailer for AnimateDiff (SEGS/pipes), and SEGSPaste in the Impact Pack, please check 1:59 in the video for how it links
Decent ❤
love u!
Good video, thanks
Thank you!
Error occurred when executing SEGSDetailerForAnimateDiff:
can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.
File "G:\Blender_ComfyUI\ComfyUI\execution.py", line 154, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "G:\Blender_ComfyUI\ComfyUI\execution.py", line 84, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
File "G:\Blender_ComfyUI\ComfyUI\execution.py", line 77, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "G:\Blender_ComfyUI\ComfyUI\custom_nodes\ComfyUI-Impact-Pack\modules\impact\segs_nodes.py", line 204, in doit
segs = SEGSDetailerForAnimateDiff.do_detail(image_frames, segs, guide_size, guide_size_for, max_size, seed, steps, cfg, sampler_name,
File "G:\Blender_ComfyUI\ComfyUI\custom_nodes\ComfyUI-Impact-Pack\modules\impact\segs_nodes.py", line 183, in do_detail
cropped_image_frames = cropped_image_frames.numpy()
😂😂😂😂
Are you able to get batch prompts/ list of prompts to work normally with text to image in comfyui like with automatic1111, I know there's the custom nodes for it but it doesn't work , you may need the only man I can ask ( que batch prompt next to que prompt is the closest kind of solution
I used ComfyUI_FizzNodes batch prompt schedule (1:19) here for inpainting, you can see both inpainting and face detailer get it, and the blinking looks good, though I haven't tried text to image yet. Check FizzNodes github, they have bunch of examples and advanced tips for customization.
How can i actually use "face detailer" node with animateddiff? It says it requires 8 latents but when i convert the image batch to image list which face detailer only accepts image lists or single images it changes it to 1 latent and causes the face to look like noise from the lacking latents.
use Simple Detector for AnimateDiff (SEGS), Detailer for AnimateDiff (SEGS/pipes), and SEGSPaste in the Impact Pack, please check 1:59 in the video for how it links