search

Naman-ntc / Pytorch-Human-Pose-Estimation

Implementation of various human pose estimation models in pytorch on multiple datasets (MPII & COCO) along with pretrained models
MIT License
472 stars 76 forks source link

Predicted result for multiple people in graph #27

Closed 54hg0220 closed 2 years ago

54hg0220 commented 2 years ago

Hi,

I have noted the MPII-Dataset labels only contains one data for each joint.

I am wondering how do you label 2 or more people on one image ? How do you represent it in the label?

Also, if possible. Could you please advise how do you evaluate your model accuracy?

Thanks

Naman-ntc commented 2 years ago

Hi, This repository is meant for single-person pose estimation only and assumes a bounding box cropped human as input. For that we use PCKh metric (although now multi-person pose estimation systems are quite common which use OKS metric)