search

SangbumChoi / MobileHumanPose

This repo is official PyTorch implementation of MobileHumanPose: Toward real-time 3D human pose estimation in mobile devices(CVPRW 2021).
MIT License
238 stars 29 forks source link

Pretrained model is not there yet? #21

Closed AivisStud closed 2 years ago

AivisStud commented 2 years ago

Dear SangbumChoi,

In repository it is said that there is pretrained model of your work. but I can not find it (I'm new to github community). Could you please guide me in the right direction where to find it or if it is not there could you share it with me. I would like to test it out and see how it works.

I managed to find ONNX version of your work in: https://github.com/PINTO0309/PINTO_model_zoo/blob/main/156_MobileHumanPose/download_mobile_human_pose_working_well.sh but I'm not sure if it works as your model and I would like to run it using pytorch. :)

Thank you for your work!

SangbumChoi commented 2 years ago

@AivisStud Thanks for asking question. Yes you can use ONNX file from that repository (I officially gave it)!. However, refactoring is still on-going due to adding various technical visualization.

AivisStud commented 2 years ago

Hi SangbumChoi,

thanks for the answer. Ok good to know. Unfortunately for me the performance was quite slow (shuttering in pc). but it could be related because of visualization scripts( i used scritps from https://github.com/ibaiGorordo/ONNX-Mobile-Human-Pose-3D and removed heatmaps). I will try to look more deeply in to the visualizations.

I will be waiting for the refactored version. Good luck with your work!:)

Thanks for your answers :)

ibaiGorordo commented 2 years ago

Sorry to interrupt here, but yes, the slow speed is probably due to the 3D plot with matplotlib. Additionally, if you want faster speed you might want to try to run it in TensorRT.