Can I use tao toolkit to improve the posenet model? (use in dusty-nv inference)

[AK51]

Hi,

I am working on a skeleton detection using dusty nv interference. It is great.
However, for the bird-eye view, the skeleton recognition is not very accurate (even if I tune down the threshold)
May I know if there is any new skeleton detection model I can try in the dusty-nv infererence? Modify the download.sh for new model download?

If not, can I train the posenet using tao toolkit? If yes, I will start to read this long manual? Is there any video tutorial?
https://docs.nvidia.com/tao/tao-toolkit/text/bodypose_estimation/bodyposenet.html

Thanks

[Morganh]

There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks

In TAO, for bodyposenet, there are two blogs. https://developer.nvidia.com/blog/training-optimizing-2d-pose-estimation-model-with-tao-toolkit-part-1/
https://developer.nvidia.com/blog/training-optimizing-2d-pose-estimation-model-with-tao-toolkit-part-2/