Thanks so much for this! What the best way to set the frame load cap? If I set too low it doesn't use the full duration of the source video and if set too high I get an error. Right now I do the math manually but I wish there was a way too get the number of frames from the source video and feed that as the frame load cap but not sure how to do this.
Hi! Awesome tutorial. But something is not working for me. I got this error from Reroute Primitive "Error occurred when executing ReroutePrimitive|pysssss: ReroutePrimitive.route() missing 1 required positional argument: 'value' File "/workspace/ComfyUI/execution.py", line 152, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) File "/workspace/ComfyUI/execution.py", line 82, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True) File "/workspace/ComfyUI/execution.py", line 70, in map_node_over_list results.append(getattr(obj, func)())"
Error occurred when executing DownloadAndLoadLivePortraitModels: Error while deserializing header: HeaderTooLarge File "D:\pinokio\api\comfyui.git\app\execution.py", line 152, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) File "D:\pinokio\api\comfyui.git\app\execution.py", line 82, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True) File "D:\pinokio\api\comfyui.git\app\execution.py", line 75, in map_node_over_list results.append(getattr(obj, func)(**slice_dict(input_data_all, i))) File "D:\pinokio\api\comfyui.git\app\custom_nodes\ComfyUI-LivePortraitKJ odes.py", line 151, in loadmodel comfy.utils.load_torch_file(feature_extractor_path) File "D:\pinokio\api\comfyui.git\app\comfy\utils.py", line 15, in load_torch_file sd = safetensors.torch.load_file(ckpt, device=device.type) File "D:\pinokio\api\comfyui.git\app\env\lib\site-packages\safetensors\torch.py", line 311, in load_file with safe_open(filename, framework="pt", device=device) as f: this error i am getting
@@CgTopTips The thing is that "Live Portrait process" that you have on your video is no longer up to date the creator himself wrote on github that something happened there. And those who updated Comfy will already have an error on this version
sorry for off-topic - but you have disabled comments on photomaker - and I want to ask something to that - in that video you loaded photomaker plus and in that node there is no 'insightface input' . when I load it in my setup, there IS, and if leave unplugged, I get an error 'Error occurred when executing PhotoMakerEncodePlus: InsightFace is required for PhotoMaker V2' - so I connect the InsightFace loader and the problem was solved ... my question is - how did you do that, that you don't need the InsightFace?
it would be helpful if you would show the end result at the BEGINNING of the video so we know a quick summary of what we're about to get into
Thanks for the feedback. You make a great point. I'll keep this in mind for future tutorials
That's right. I always look for the end result at the BEGINNING of the video to avoid unnecessary waste of time for wrong tutorial.
Thanks so much for this! What the best way to set the frame load cap? If I set too low it doesn't use the full duration of the source video and if set too high I get an error. Right now I do the math manually but I wish there was a way too get the number of frames from the source video and feed that as the frame load cap but not sure how to do this.
Hi! Awesome tutorial. But something is not working for me. I got this error from Reroute Primitive
"Error occurred when executing ReroutePrimitive|pysssss:
ReroutePrimitive.route() missing 1 required positional argument: 'value'
File "/workspace/ComfyUI/execution.py", line 152, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "/workspace/ComfyUI/execution.py", line 82, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
File "/workspace/ComfyUI/execution.py", line 70, in map_node_over_list
results.append(getattr(obj, func)())"
Error occurred when executing DownloadAndLoadLivePortraitModels:
Error while deserializing header: HeaderTooLarge
File "D:\pinokio\api\comfyui.git\app\execution.py", line 152, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "D:\pinokio\api\comfyui.git\app\execution.py", line 82, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
File "D:\pinokio\api\comfyui.git\app\execution.py", line 75, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "D:\pinokio\api\comfyui.git\app\custom_nodes\ComfyUI-LivePortraitKJ
odes.py", line 151, in loadmodel
comfy.utils.load_torch_file(feature_extractor_path)
File "D:\pinokio\api\comfyui.git\app\comfy\utils.py", line 15, in load_torch_file
sd = safetensors.torch.load_file(ckpt, device=device.type)
File "D:\pinokio\api\comfyui.git\app\env\lib\site-packages\safetensors\torch.py", line 311, in load_file
with safe_open(filename, framework="pt", device=device) as f:
this error i am getting
Hello, thank you for your work. How to save a finished material separately. When saving, two images are saved at once
Delete "Image Concatenate Multi" node and connect "Get image size & count" node directly to "video combine" node
excellent! how to modify this to image2video? can you replace the first node with image?
Please watch this video:
ruclips.net/video/yEjnAV9AOxQ/видео.html
@@CgTopTips The thing is that "Live Portrait process" that you have on your video is no longer up to date the creator himself wrote on github that something happened there. And those who updated Comfy will already have an error on this version
你好，例子中生成后的视频手部的地方会发生闪烁或模糊，这个是在哪里控制？
Hello and thanks, you don't need to conncet node with blazeface?
You are right. Sorry, I forgot to connect it. BlazeFace is the best choice for face detection !
Thanks 🎉😊
sorry for off-topic - but you have disabled comments on photomaker - and I want to ask something to that - in that video you loaded photomaker plus and in that node there is no 'insightface input' . when I load it in my setup, there IS, and if leave unplugged, I get an error 'Error occurred when executing PhotoMakerEncodePlus: InsightFace is required for PhotoMaker V2' - so I connect the InsightFace loader and the problem was solved ... my question is - how did you do that, that you don't need the InsightFace?
720P?
I forgot to set the screen recorder setting on FHD :(
Great 👍