Home

Awesome

POTTER: Pooling Attention Transformer for Efficient Human Mesh Recovery

The project is an official implementation of our paper POTTER: Pooling Attention Transformer for Efficient Human Mesh Recovery, CVPR 2023.

News :triangular_flag_on_post:

Please check image_classification folder for Image Classification Part (Section 4.1 in our paper) and human_mesh_recovery folder for Human Mesh Recovery Part.

Experiments

<p align="center"> <img src="./img/w1.png"> <img src="./img/w2.png"> </p>

Visualizations

<p align="center"> <img src="./img/w3.png"> </p>

Citing

If our code helps your research, please consider citing the following paper:

@inproceedings{zheng2023potter,
    title={POTTER: Pooling Attention Transformer for Efficient Human Mesh Recovery},
    author={Zheng, Ce and Liu, Xianpeng and Qi, Guo-Jun and Chen, Chen},
    booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
    year={2023}
}

@inproceedings{zheng2023feater,
    title={FeatER: An Efficient Network for Human Reconstruction via Feature Map-Based TransformER},
    author={Zheng, Ce and Mendieta, Matias and Yang, Taojiannan and Qi, Guo-Jun and Chen, Chen},
    booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
    year={2023}
}