HRNet (CVPR'2019)
```bibtex
@inproceedings{sun2019deep,
title={Deep high-resolution representation learning for human pose estimation},
author={Sun, Ke and Xiao, Bin and Liu, Dong and Wang, Jingdong},
booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
pages={5693--5703},
year={2019}
}
```
UDP (CVPR'2020)
```bibtex
@InProceedings{Huang_2020_CVPR,
author = {Huang, Junjie and Zhu, Zheng and Guo, Feng and Huang, Guan},
title = {The Devil Is in the Details: Delving Into Unbiased Data Processing for Human Pose Estimation},
booktitle = {The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2020}
}
```
COCO (ECCV'2014)
```bibtex
@inproceedings{lin2014microsoft,
title={Microsoft coco: Common objects in context},
author={Lin, Tsung-Yi and Maire, Michael and Belongie, Serge and Hays, James and Perona, Pietro and Ramanan, Deva and Doll{\'a}r, Piotr and Zitnick, C Lawrence},
booktitle={European conference on computer vision},
pages={740--755},
year={2014},
organization={Springer}
}
```
Results on COCO val2017 with detector having human AP of 56.4 on COCO val2017 dataset
| Arch | Input Size | AP | AP50 | AP75 | AR | AR50 | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [pose_hrnet_w32_udp](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_hrnet-w32_udp-8xb64-210e_coco-256x192.py) | 256x192 | 0.762 | 0.907 | 0.829 | 0.810 | 0.942 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_hrnet-w32_udp-8xb64-210e_coco-256x192-73ede547_20220914.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_hrnet-w32_udp-8xb64-210e_coco-256x192_20220914.log) |
| [pose_hrnet_w32_udp](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_hrnet-w32_udp-8xb64-210e_coco-384x288.py) | 384x288 | 0.768 | 0.909 | 0.832 | 0.815 | 0.945 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_hrnet-w32_udp-8xb64-210e_coco-384x288-9a3f7c85_20220914.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_hrnet-w32_udp-8xb64-210e_coco-384x288_20220914.log) |
| [pose_hrnet_w48_udp](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_hrnet-w48_udp-8xb32-210e_coco-256x192.py) | 256x192 | 0.768 | 0.908 | 0.833 | 0.817 | 0.945 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_hrnet-w48_udp-8xb32-210e_coco-256x192-3feaef8f_20220913.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_hrnet-w48_udp-8xb32-210e_coco-256x192_20220913.log) |
| [pose_hrnet_w48_udp](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_hrnet-w48_udp-8xb32-210e_coco-384x288.py) | 384x288 | 0.773 | 0.911 | 0.836 | 0.821 | 0.946 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_hrnet-w48_udp-8xb32-210e_coco-384x288-70d7ab01_20220913.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_hrnet-w48_udp-8xb32-210e_coco-384x288_20220913.log) |
| [pose_hrnet_w32_udp_regress](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_hrnet-w32_udp-regress-8xb64-210e_coco-256x192.py) | 256x192 | 0.759 | 0.907 | 0.827 | 0.813 | 0.943 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_hrnet-w32_udp-regress-8xb64-210e_coco-256x192-9c0b77b4_20220926.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_hrnet-w32_udp-regress-8xb64-210e_coco-256x192_20220226.log) |
Note that, UDP also adopts the unbiased encoding/decoding algorithm of [DARK](https://mmpose.readthedocs.io/en/latest/model_zoo_papers/techniques.html#darkpose-cvpr-2020).