ComfyUI-Impact-Pack/Inspire-Pack - MediaPIpe FaceMesh Detector

This video introduces the detection capabilities based on the MediaPipe FaceMesh from the Impact Pack and Inspire Pack.

[bzikarius]

You are my hero! I tried hard to find info about SAM automasking and finally found it here!

[AIMusicExperiment]

This advancement has revolutionized my workflows! I cannot overstate how groundbreaking this is! Thanks @drltdata

[sebaccimaster]

This channel has the juice! Thanks for sharing

[romanbulatnikov]

This is really amazing technique to work on the eyes, thank you for sharing👍 In my work I am frequently having different eye color / tint when processing, do you think there is some way to better control it ?

[StargateMax]

I'm unable to find the node anywhere, I did install the addon packs.

[MrLinX]

I have this Error :
Error occurred when executing

name 'framework' is not defined

File "C:\ComfyUI\ComfyUI_windows_portable\ComfyUI\execution.py", line 152, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "C:\ComfyUI\ComfyUI_windows_portable\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 "C:\ComfyUI\ComfyUI_windows_portable\ComfyUI\execution.py", line 75, 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_controlnet_aux\node_wrappers\mediapipe_face.py", line 41, in detect
import mediapipe
File "C:\ComfyUI\ComfyUI_windows_portable\python_embeded\lib\site-packages\mediapipe\__init__.py", line 20, in <module>
del framework

Impact Pack, Inspire Pack and Controlnet Auxiliary Preprocessors are up-to-date. No missing custom node.

[masa]

Thank you so much for your instruction!
I've successful to make first part to make Facemesh SEG but to enable Face detailer(pipe) node I've got this error.
I placed control_v2p_sd15_mediapipe_face.safetensors, control_v2p_sd15_mediapipe_face.pth, into Controlnet folder.
Could you please tell me what am I still missing?

Error occurred when executing FaceDetailerPipe:

got an unexpected keyword argument 'resolution'

File "D:\Python\path\my_venv\ComfyUI\ComfyUI_windows_portable_nvidia_cu118_or_cpu\ComfyUI_windows_portable\ComfyUI\execution.py", line 153, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "D:\Python\path\my_venv\ComfyUI\ComfyUI_windows_portable_nvidia_cu118_or_cpu\ComfyUI_windows_portable\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:\Python\path\my_venv\ComfyU\ComfyUI_windows_portable_nvidia_cu118_or_cpu\ComfyUI_windows_portable\ComfyUI\execution.py", line 76, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "D:\Python\path\my_venv\ComfyUI\ComfyUI_windows_portable_nvidia_cu118_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Impact-Pack\modules\impact\impact_pack.py", line 978, in doit
enhanced_img, cropped_enhanced, cropped_enhanced_alpha, mask, cnet_pil_list = FaceDetailer.enhance_face(
File "D:\Python\path\my_venv\ComfyUI\ComfyUI_windows_portable_nvidia_cu118_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Impact-Pack\modules\impact\impact_pack.py", line 376, in enhance_face
segs = bbox_detector.detect(image, bbox_threshold, bbox_dilation, bbox_crop_factor, drop_size)
File "D:\Python\path\my_venv\ComfyUI\ComfyUI_windows_portable_nvidia_cu118_or_cpu\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Inspire-Pack\segs_support.py", line 36, in detect
facemesh_image = pre_obj().detect(image, self.max_faces, threshold, resolution=resolution)[0]

[vincentdutour]

Hi ! How could i process an image list as input with your workflow ? Work perfectly for one image but when the input is a video, the media pipe face mesh wont process a list of images, do you have some trick for me ? thanks :)

[vincentdutour]

Hi ! Got an error : Error occurred when executing MediaPipeFaceMeshToSEGS:

OpenCV(4.8.1) error: (-215:Assertion failed) top >= 0 && bottom >= 0 && left >= 0 && right >= 0 && _src.dims() <= 2 in function 'cv::copyMakeBorder'

All is up to date, how could i resolve this problem ?
Best

[drltdata]

NOTE: When the seed widget of each node is not set to 'fixed, ' GlobalSeed's value is applied correctly, but on the screen, it is overridden by 'randomize', making the applied seed not visible. However, it is reflected when you load the generated workflow.

You can download from ComfyUI from here:

And ComfyUI Inspire Pack extension from here:

And ComfyUI's ControlNet Auxiliary Preprocessors extension from here:

And workflow from here (you can drag & drop):

[luislozano]

How do you get the names on the right side of node with the author name? What do i have to install or what setting do i have to turn on? This is great for helping to identify similar nodes!

[EITCHS]

can anyone advise when im using the node, im getting expected keyword argument ‘resolution’?

[marcihuppi]

Error occurred when executing FaceDetailerPipe:

got an unexpected keyword argument 'resolution'


I always get this error :(

[AiZill]

I get an error: Error occurred when executing FaceDetailer:

got an unexpected keyword argument 'resolution'

File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 151, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "D:\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 "D:\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 "D:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Impact-Pack\modules\impact\impact_pack.py", line 416, in doit
enhanced_img, cropped_enhanced, cropped_enhanced_alpha, mask, cnet_pil_list = FaceDetailer.enhance_face(
File "D:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Impact-Pack\modules\impact\impact_pack.py", line 373, in enhance_face
segs = bbox_detector.detect(image, bbox_threshold, bbox_dilation, bbox_crop_factor, drop_size)
File "D:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Inspire-Pack\segs_support.py", line 35, in detect
facemesh_image = pre_obj().detect(image, self.max_faces, threshold, resolution=resolution)[0]