search

paTRICK-swk / P-STMO

[ECCV2022] The PyTorch implementation for "P-STMO: Pre-Trained Spatial Temporal Many-to-One Model for 3D Human Pose Estimation"
MIT License
154 stars 11 forks source link

Using refined model #1

Closed cbenitez81 closed 2 years ago

cbenitez81 commented 2 years ago

Hi thanks for the paper to start with.

I wondering about the refined model, it seems that needs the 3d ground truth to make the refinement and the camera parameters?. I was wondering if you obtained those automatically or it's kind of overfitting to the 3d ground-truth in some way

thanks

paTRICK-swk commented 2 years ago

You're right. It does lead to some degree of overfitting. On the other hand, the 3d ground-truth and camera parameters are not available when testing on custom videos, so the refining module does not work.

vicentowang commented 2 years ago

You're right. It does lead to some degree of overfitting. On the other hand, the 3d ground-truth and camera parameters are not available when testing on custom videos, so the refining module does not work.

may it be possiable to use DNN( inpust_2d ) to predict a camera parameters for the wild video ? camera parameters refer to f and c ?