Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

continuous message #109

Open
Ishihara-Masabumi opened this issue Oct 4, 2023 · 3 comments
Open

continuous message #109

Ishihara-Masabumi opened this issue Oct 4, 2023 · 3 comments
Assignees
Labels
question Further information is requested

Comments

@Ishihara-Masabumi
Copy link

Ishihara-Masabumi commented Oct 4, 2023

At the following loop,

while True:
    try:
        image = take_frame()

        image = resize_with_padding(image,1280,720)

        # Perform image processing with MediaPipeHolistic
        mocapNETInput, annotated_image = mp_holistic.convertImageToMocapNETInput(image)

        # Perform 3D joint prediction with MocapNET
        mocapNET3DOutput = mnet.predict3DJoints(mocapNETInput)

        # Visualize the results on the image
        #image_with_results = Image.fromarray(cv2.cvtColor(annotated_image, cv2.COLOR_BGR2RGB))

        from MocapNETVisualization import visualizeMocapNETEnsemble
        image,plotImage = visualizeMocapNETEnsemble(mnet,annotated_image,plotBVHChannels=False)

        show_frame(annotated_image)  # it replace previous image

    except Exception as err:
        print('Exception:', err)

there is the continuous message as the following example.

prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element
is missing
2dx_endsite_toe5-3.r is missing 2dy_endsite_toe5-3.r` is missing visible_endsite_toe5-3.r is missing 2dx_lhip is missing 2dy_lhip is missing visible_lhip 2dx_lknee is missing 2dy_lknee is missing visible_lknee 2dx_lfoot is missing 2dy_lfoot is missing visible_lfoot is missing 2dx_endsite_toel-2.1
is missing
visible_endsite_toel-2.l
is missing
2dy_endsite_toe1-2.L` is missing
prepareInputG: NSRM Input Element
is missing
prepareInputG: NSRM Input Element
2dx_endsite_toe5-3.1 is missing
prepareInputG: NSRM Input Element
prepareInputG: NSRM Input Element
is missing
2dy_endsite_toe5-3.L` is missing visible_endsite_toe5-3.1
Too many elements missing for lower body missingRatio: 1.0..
prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element prepareInputG: NSRM Input Element
2dx_hip is missing 2dy_hip is missing visible hip is missing
2dx_neck1 is missing 2dy_neck1 is missing visible_neck1 is missing

Is that OK?

@AmmarkoV AmmarkoV self-assigned this Oct 8, 2023
@AmmarkoV AmmarkoV added the question Further information is requested label Oct 8, 2023
@AmmarkoV
Copy link
Collaborator

AmmarkoV commented Oct 8, 2023

This warning message indicates that the toes are missing.
This of course has a slightly negative impact on the neural network regression (since it is missing part of the input normally required for full operation)

The network formulation is based on the BODY25 standard, when using the network with an RGB->2D joint estimator that does not have toes, the network can however still operate and produce results, so it is OK.

49032982-d5bfb580-f162-11e8-935d-5139c01eeb57

@Ishihara-Masabumi
Copy link
Author

Ishihara-Masabumi commented Oct 19, 2023

This time, I have the similar messages over 15 hours on the mocapnet4.ipynb as follows.

 prepareInputG: NSRM Input Element ` visible_lfoot ` is missing 
 prepareInputG: NSRM Input Element ` 2dx_endsite_toe1-2.l ` is missing 
 prepareInputG: NSRM Input Element ` 2dy_endsite_toe1-2.l ` is missing 
 prepareInputG: NSRM Input Element ` visible_endsite_toe1-2.l ` is missing 
 prepareInputG: NSRM Input Element ` 2dx_endsite_toe5-3.l ` is missing 
 prepareInputG: NSRM Input Element ` 2dy_endsite_toe5-3.l ` is missing 
 prepareInputG: NSRM Input Element ` visible_endsite_toe5-3.l ` is missing 
 Too many elements missing for  lowerbody  missingRatio :  1.0 .. 
 prepareInputG: NSRM Input Element ` 2dx_hip ` is missing 
 prepareInputG: NSRM Input Element ` 2dy_hip ` is missing 
 prepareInputG: NSRM Input Element ` visible_hip ` is missing 
 prepareInputG: NSRM Input Element ` 2dx_neck1 ` is missing 
 prepareInputG: NSRM Input Element ` 2dy_neck1 ` is missing 
 prepareInputG: NSRM Input Element ` visible_neck1 ` is missing 
 prepareInputG: NSRM Input Element ` 2dx_head ` is missing 
 prepareInputG: NSRM Input Element ` 2dy_head ` is missing 
 prepareInputG: NSRM Input Element ` visible_head ` is missing 
 prepareInputG: NSRM Input Element ` 2dx_endsite_eye.l ` is missing 
 prepareInputG: NSRM Input Element ` 2dy_endsite_eye.l ` is missing 
 prepareInputG: NSRM Input Element ` visible_endsite_eye.l ` is missing 
 prepareInputG: NSRM Input Element ` 2dx_endsite_eye.r ` is missing 
 prepareInputG: NSRM Input Element ` 2dy_endsite_eye.r ` is missing 
 prepareInputG: NSRM Input Element ` visible_endsite_eye.r ` is missing 
 prepareInputG: NSRM Input Element ` 2dx_rshoulder ` is missing 
 prepareInputG: NSRM Input Element ` 2dy_rshoulder ` is missing 
 prepareInputG: NSRM Input Element ` visible_rshoulder ` is missing 
 prepareInputG: NSRM Input Element ` 2dx_relbow ` is missing 
 prepareInputG: NSRM Input Element ` 2dy_relbow ` is missing 
 prepareInputG: NSRM Input Element ` visible_relbow ` is missing 
 prepareInputG: NSRM Input Element ` 2dx_rhand ` is missing 
 prepareInputG: NSRM Input Element ` 2dy_rhand ` is missing 
 prepareInputG: NSRM Input Element ` visible_rhand ` is missing 
 prepareInputG: NSRM Input Element ` 2dx_lshoulder ` is missing 
 prepareInputG: NSRM Input Element ` 2dy_lshoulder ` is missing 
 prepareInputG: NSRM Input Element ` visible_lshoulder ` is missing 
 prepareInputG: NSRM Input Element ` 2dx_lelbow ` is missing 
 prepareInputG: NSRM Input Element ` 2dy_lelbow ` is missing 
 prepareInputG: NSRM Input Element ` visible_lelbow ` is missing 
 prepareInputG: NSRM Input Element ` 2dx_lhand ` is missing 
 prepareInputG: NSRM Input Element ` 2dy_lhand ` is missing 
 prepareInputG: NSRM Input Element ` visible_lhand ` is missing 
 Too many elements missing for  upperbody  missingRatio :  1.0 .. 
...
...
...
 prepareInputG: NSRM Input Element ` 2dy_rhand ` is missing 
 prepareInputG: NSRM Input Element ` visible_rhand ` is missing 
 prepareInputG: NSRM Input Element ` 2dx_lshoulder ` is missing 
 prepareInputG: NSRM Input Element ` 2dy_lshoulder ` is missing 
 prepareInputG: NSRM Input Element ` visible_lshoulder ` is missing 
 prepareInputG: NSRM Input Element ` 2dx_lelbow ` is missing 
 prepareInputG: NSRM Input Element ` 2dy_lelbow ` is missing 
 prepareInputG: NSRM Input Element ` visible_lelbow ` is missing 
 prepareInputG: NSRM Input Element ` 2dx_lhand ` is missing 
 prepareInputG: NSRM Input Element ` 2dy_lhand ` is missing 
 prepareInputG: NSRM Input Element ` visible_lhand ` is missing 
 Too many elements missing for  upperbody  missingRatio :  1.0 .. 

Is this situation OK, or NG?

@AmmarkoV
Copy link
Collaborator

AmmarkoV commented Dec 8, 2023

Once again this means that the 2D pose estimation module (mediapipe/openpose/etc) could not find the 2D joint location from the RGB image. Of course without their 2D location MocapNET will not be able to correctly answer about their 3D pose, and this is the reason for logging them, however it is normal to happen so OK I guess

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
question Further information is requested
Projects
None yet
Development

No branches or pull requests

2 participants