Pose estimation: I am getting NaN for all the keypoints aswell as the score about tfjs

[autotel]

I found a solution that works for myself: I changed the detection model from BlazePose into MoveNet

simply changed the statement

const detector = await poseDetection.createDetector(
    poseDetection.SupportedModels.BlazePose,
    {
        runtime: 'tfjs',
        enableSmoothing: true,

    });

into

const detector = await poseDetection.createDetector(
    poseDetection.SupportedModels.MoveNet,
    {
        runtime: 'tfjs',
        enableSmoothing: true,

    });

and It works.
Still curious as to why BlazePose would output only NaNs without throwing any error.

[gaikwadrahul8]

Hi, @autotel

I apologize for the delayed response and good to hear that your issue got resolved after changing poseDetection.SupportedModels.BlazePose to poseDetection.SupportedModels.MoveNet, actually I'm trying to replicate the same behaviour from my end but after running npm run serve command I'm getting DONE Compiled successfully in 1414ms message when I'm trying run http://localhost:8080/?model=blazepose and http://localhost:8080/?model=MoveNet after clicking on Start Capture button I'm getting below error message even though I have given permission to my camera so do you have any idea why I'm getting below error, if so please guide me so I can replicate same behavior from my end also.

Here is error screenshot for reference :

Thank you for your cooperation and patience.

[github-actions]

This issue has been marked stale because it has no recent activity since 7 days. It will be closed if no further activity occurs. Thank you.

[github-actions]

This issue was closed due to lack of activity after being marked stale for past 7 days.

[google-ml-butler]

Are you satisfied with the resolution of your issue?
Yes
No