Skip to content

Pytorch code for Hybrid Coarse-fine Classification for Head Pose Estimation

Notifications You must be signed in to change notification settings

wqz960/accurate-head-pose

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

11 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

accurate-head-pose

We release the code of the Hybrid Coarse-fine Classification for Head Pose Estimation, built on top of the deep-head-pose.

Pretrained model

We provide pretrained model to reproduce the same result shown in the paper.

AFLW2000, password: drmz

AFLW, password: yym5

BIWI, password: 8qpc

For those who cannot have access to BaiduDisk, you can download pretrained models on Google Drive

Testing

Training and testing lists can be found in /tools, you need download corresonding dataset and update the path.
AFLW2000 dataset, password: xr6e

python test_hopenet.py --gpu 0 --data_dir directory-path-for-dataset --filename_list filename-list --snapshot model --dataset dataset-name 

TODO

Instructions for scripts
Better and better models
Videos and example demo

Cite this work

Haofan Wang, Zhenhua Chen and Yi Zhou "Hybrid coarse-fine classification for head pose estimation." arXiv:1901.06778, 2019. (Download)

Biblatex entry:

        @article{wang2019hybrid,
          title={Hybrid coarse-fine classification for head pose estimation},
          author={Wang, Haofan and Chen, Zhenghua and Zhou, Yi},
          journal={arXiv preprint arXiv:1901.06778},
          year={2019}
        }

Acknowledgement

Our hybrid classification network is plug-and-play on top of the deep-head-pose, but it could be extended to other classification tasks easily. We thank Nataniel Ruiz for releasing deep-head-pose-Pytorch codebase.

About

Pytorch code for Hybrid Coarse-fine Classification for Head Pose Estimation

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%