Commit ca8a762a authored by chenzk's avatar chenzk
Browse files

v1.0

parents
Pipeline #716 failed with stages
in 0 seconds
<!-- [BACKBONE] -->
<details>
<summary align="right"><a href="https://arxiv.org/abs/2102.12122">PVT (ICCV'2021)</a></summary>
```bibtex
@inproceedings{wang2021pyramid,
title={Pyramid vision transformer: A versatile backbone for dense prediction without convolutions},
author={Wang, Wenhai and Xie, Enze and Li, Xiang and Fan, Deng-Ping and Song, Kaitao and Liang, Ding and Lu, Tong and Luo, Ping and Shao, Ling},
booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision},
pages={568--578},
year={2021}
}
```
</details>
<details>
<summary align="right"><a href="https://arxiv.org/abs/2106.13797">PVTV2 (CVMJ'2022)</a></summary>
```bibtex
@article{wang2022pvt,
title={PVT v2: Improved baselines with Pyramid Vision Transformer},
author={Wang, Wenhai and Xie, Enze and Li, Xiang and Fan, Deng-Ping and Song, Kaitao and Liang, Ding and Lu, Tong and Luo, Ping and Shao, Ling},
journal={Computational Visual Media},
pages={1--10},
year={2022},
publisher={Springer}
}
```
</details>
<!-- [DATASET] -->
<details>
<summary align="right"><a href="https://link.springer.com/chapter/10.1007/978-3-319-10602-1_48">COCO (ECCV'2014)</a></summary>
```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}
}
```
</details>
Results on COCO val2017 with detector having human AP of 56.4 on COCO val2017 dataset
| Arch | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [pose_pvt-s](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_pvt-s_8xb64-210e_coco-256x192.py) | 256x192 | 0.714 | 0.896 | 0.794 | 0.773 | 0.936 | [ckpt](https://download.openmmlab.com/mmpose/top_down/pvt/pvt_small_coco_256x192-4324a49d_20220501.pth) | [log](https://download.openmmlab.com/mmpose/top_down/pvt/pvt_small_coco_256x192_20220501.log.json) |
| [pose_pvtv2-b2](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_pvtv2-b2_8xb64-210e_coco-256x192.py) | 256x192 | 0.737 | 0.905 | 0.812 | 0.791 | 0.942 | [ckpt](https://download.openmmlab.com/mmpose/top_down/pvt/pvtv2_b2_coco_256x192-b4212737_20220501.pth) | [log](https://download.openmmlab.com/mmpose/top_down/pvt/pvtv2_b2_coco_256x192_20220501.log.json) |
Models:
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_pvt-s_8xb64-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: &id001
- SimpleBaseline2D
- PVT
Training Data: COCO
Name: td-hm_pvt-s_8xb64-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.714
AP@0.5: 0.896
AP@0.75: 0.794
AR: 0.773
AR@0.5: 0.936
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/pvt/pvt_small_coco_256x192-4324a49d_20220501.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_pvtv2-b2_8xb64-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_pvtv2-b2_8xb64-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.737
AP@0.5: 0.905
AP@0.75: 0.812
AR: 0.791
AR@0.5: 0.942
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/pvt/pvtv2_b2_coco_256x192-b4212737_20220501.pth
<!-- [BACKBONE] -->
<details>
<summary align="right"><a href="https://arxiv.org/abs/2004.08955">ResNeSt (ArXiv'2020)</a></summary>
```bibtex
@article{zhang2020resnest,
title={ResNeSt: Split-Attention Networks},
author={Zhang, Hang and Wu, Chongruo and Zhang, Zhongyue and Zhu, Yi and Zhang, Zhi and Lin, Haibin and Sun, Yue and He, Tong and Muller, Jonas and Manmatha, R. and Li, Mu and Smola, Alexander},
journal={arXiv preprint arXiv:2004.08955},
year={2020}
}
```
</details>
<!-- [DATASET] -->
<details>
<summary align="right"><a href="https://link.springer.com/chapter/10.1007/978-3-319-10602-1_48">COCO (ECCV'2014)</a></summary>
```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}
}
```
</details>
Results on COCO val2017 with detector having human AP of 56.4 on COCO val2017 dataset
| Arch | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [pose_resnest_50](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnest50_8xb64-210e_coco-256x192.py) | 256x192 | 0.720 | 0.899 | 0.800 | 0.775 | 0.939 | [ckpt](https://download.openmmlab.com/mmpose/top_down/resnest/resnest50_coco_256x192-6e65eece_20210320.pth) | [log](https://download.openmmlab.com/mmpose/top_down/resnest/resnest50_coco_256x192_20210320.log.json) |
| [pose_resnest_50](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnest50_8xb64-210e_coco-384x288.py) | 384x288 | 0.737 | 0.900 | 0.811 | 0.789 | 0.937 | [ckpt](https://download.openmmlab.com/mmpose/top_down/resnest/resnest50_coco_384x288-dcd20436_20210320.pth) | [log](https://download.openmmlab.com/mmpose/top_down/resnest/resnest50_coco_384x288_20210320.log.json) |
| [pose_resnest_101](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnest101_8xb64-210e_coco-256x192.py) | 256x192 | 0.725 | 0.900 | 0.807 | 0.781 | 0.939 | [ckpt](https://download.openmmlab.com/mmpose/top_down/resnest/resnest101_coco_256x192-2ffcdc9d_20210320.pth) | [log](https://download.openmmlab.com/mmpose/top_down/resnest/resnest101_coco_256x192_20210320.log.json) |
| [pose_resnest_101](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnest101_8xb32-210e_coco-384x288.py) | 384x288 | 0.745 | 0.905 | 0.818 | 0.798 | 0.942 | [ckpt](https://download.openmmlab.com/mmpose/top_down/resnest/resnest101_coco_384x288-80660658_20210320.pth) | [log](https://download.openmmlab.com/mmpose/top_down/resnest/resnest101_coco_384x288_20210320.log.json) |
| [pose_resnest_200](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnest200_8xb64-210e_coco-256x192.py) | 256x192 | 0.731 | 0.905 | 0.812 | 0.787 | 0.943 | [ckpt](https://download.openmmlab.com/mmpose/top_down/resnest/resnest200_coco_256x192-db007a48_20210517.pth) | [log](https://download.openmmlab.com/mmpose/top_down/resnest/resnest200_coco_256x192_20210517.log.json) |
| [pose_resnest_200](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnest200_8xb16-210e_coco-384x288.py) | 384x288 | 0.753 | 0.907 | 0.827 | 0.805 | 0.943 | [ckpt](https://download.openmmlab.com/mmpose/top_down/resnest/resnest200_coco_384x288-b5bb76cb_20210517.pth) | [log](https://download.openmmlab.com/mmpose/top_down/resnest/resnest200_coco_384x288_20210517.log.json) |
| [pose_resnest_269](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnest269_8xb32-210e_coco-256x192.py) | 256x192 | 0.737 | 0.907 | 0.819 | 0.792 | 0.943 | [ckpt](https://download.openmmlab.com/mmpose/top_down/resnest/resnest269_coco_256x192-2a7882ac_20210517.pth) | [log](https://download.openmmlab.com/mmpose/top_down/resnest/resnest269_coco_256x192_20210517.log.json) |
| [pose_resnest_269](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnest269_8xb16-210e_coco-384x288.py) | 384x288 | 0.754 | 0.908 | 0.828 | 0.805 | 0.943 | [ckpt](https://download.openmmlab.com/mmpose/top_down/resnest/resnest269_coco_384x288-b142b9fb_20210517.pth) | [log](https://download.openmmlab.com/mmpose/top_down/resnest/resnest269_coco_384x288_20210517.log.json) |
Models:
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnest50_8xb64-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: &id001
- SimpleBaseline2D
- ResNeSt
Training Data: COCO
Name: td-hm_resnest50_8xb64-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.720
AP@0.5: 0.899
AP@0.75: 0.8
AR: 0.775
AR@0.5: 0.939
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/resnest/resnest50_coco_256x192-6e65eece_20210320.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnest50_8xb64-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_resnest50_8xb64-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.737
AP@0.5: 0.9
AP@0.75: 0.811
AR: 0.789
AR@0.5: 0.937
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/resnest/resnest50_coco_384x288-dcd20436_20210320.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnest101_8xb64-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_resnest101_8xb64-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.725
AP@0.5: 0.9
AP@0.75: 0.807
AR: 0.781
AR@0.5: 0.939
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/resnest/resnest101_coco_256x192-2ffcdc9d_20210320.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnest101_8xb32-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_resnest101_8xb32-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.745
AP@0.5: 0.905
AP@0.75: 0.818
AR: 0.798
AR@0.5: 0.942
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/resnest/resnest101_coco_384x288-80660658_20210320.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnest200_8xb64-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_resnest200_8xb64-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.731
AP@0.5: 0.905
AP@0.75: 0.812
AR: 0.787
AR@0.5: 0.943
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/resnest/resnest200_coco_256x192-db007a48_20210517.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnest200_8xb16-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_resnest200_8xb16-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.753
AP@0.5: 0.907
AP@0.75: 0.827
AR: 0.805
AR@0.5: 0.943
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/resnest/resnest200_coco_384x288-b5bb76cb_20210517.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnest269_8xb32-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_resnest269_8xb32-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.737
AP@0.5: 0.907
AP@0.75: 0.819
AR: 0.792
AR@0.5: 0.943
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/resnest/resnest269_coco_256x192-2a7882ac_20210517.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnest269_8xb16-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_resnest269_8xb16-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.754
AP@0.5: 0.908
AP@0.75: 0.828
AR: 0.805
AR@0.5: 0.943
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/resnest/resnest269_coco_384x288-b142b9fb_20210517.pth
<!-- [ALGORITHM] -->
<details>
<summary align="right"><a href="http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html">SimpleBaseline2D (ECCV'2018)</a></summary>
```bibtex
@inproceedings{xiao2018simple,
title={Simple baselines for human pose estimation and tracking},
author={Xiao, Bin and Wu, Haiping and Wei, Yichen},
booktitle={Proceedings of the European conference on computer vision (ECCV)},
pages={466--481},
year={2018}
}
```
</details>
<!-- [BACKBONE] -->
<details>
<summary align="right"><a href="http://openaccess.thecvf.com/content_cvpr_2016/html/He_Deep_Residual_Learning_CVPR_2016_paper.html">ResNet (CVPR'2016)</a></summary>
```bibtex
@inproceedings{he2016deep,
title={Deep residual learning for image recognition},
author={He, Kaiming and Zhang, Xiangyu and Ren, Shaoqing and Sun, Jian},
booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
pages={770--778},
year={2016}
}
```
</details>
<!-- [DATASET] -->
<details>
<summary align="right"><a href="https://link.springer.com/chapter/10.1007/978-3-319-10602-1_48">COCO (ECCV'2014)</a></summary>
```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}
}
```
</details>
Results on COCO val2017 with detector having human AP of 56.4 on COCO val2017 dataset
| Arch | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [pose_resnet_50](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_8xb64-210e_coco-256x192.py) | 256x192 | 0.718 | 0.898 | 0.796 | 0.774 | 0.934 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_8xb64-210e_coco-256x192-04af38ce_20220923.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_8xb64-210e_coco-256x192_20220923.log) |
| [pose_resnet_50](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_8xb64-210e_coco-384x288.py) | 384x288 | 0.731 | 0.900 | 0.799 | 0.782 | 0.937 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_8xb64-210e_coco-384x288-7b8db90e_20220923.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_8xb64-210e_coco-384x288_20220923.log) |
| [pose_resnet_101](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_8xb64-210e_coco-256x192.py) | 256x192 | 0.728 | 0.904 | 0.809 | 0.783 | 0.942 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_8xb64-210e_coco-256x192-065d3625_20220926.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_8xb64-210e_coco-256x192_20220926.log) |
| [pose_resnet_101](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_8xb32-210e_coco-384x288.py) | 384x288 | 0.749 | 0.906 | 0.817 | 0.799 | 0.941 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_8xb64-210e_coco-256x192-065d3625_20220926.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_8xb64-210e_coco-256x192_20220926.log) |
| [pose_resnet_152](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_8xb32-210e_coco-256x192.py) | 256x192 | 0.736 | 0.904 | 0.818 | 0.791 | 0.942 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_8xb32-210e_coco-256x192-0345f330_20220928.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_8xb32-210e_coco-256x192_20220928.log) |
| [pose_resnet_152](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_8xb32-210e_coco-384x288.py) | 384x288 | 0.750 | 0.908 | 0.821 | 0.800 | 0.942 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_8xb32-210e_coco-384x288-7fbb906f_20220927.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_8xb32-210e_coco-384x288_20220927.log) |
The following model is equipped with a visibility prediction head and has been trained using COCO and AIC datasets.
| Arch | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [pose_resnet_50](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm-vis_res50_8xb64-210e_coco-aic-256x192-merge.py) | 256x192 | 0.729 | 0.900 | 0.807 | 0.783 | 0.938 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm-vis_res50_8xb64-210e_coco-aic-256x192-merge-21815b2c_20230726.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_8xb64-210e_coco-256x192_20220923.log) |
Collections:
- Name: SimpleBaseline2D
Paper:
Title: Simple baselines for human pose estimation and tracking
URL: http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html
README: https://github.com/open-mmlab/mmpose/blob/main/docs/src/papers/algorithms/simplebaseline2d.md
Models:
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_8xb64-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: &id001
- SimpleBaseline2D
- ResNet
Training Data: COCO
Name: td-hm_res50_8xb64-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.718
AP@0.5: 0.898
AP@0.75: 0.796
AR: 0.774
AR@0.5: 0.934
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_8xb64-210e_coco-256x192-04af38ce_20220923.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_8xb64-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_res50_8xb64-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.731
AP@0.5: 0.9
AP@0.75: 0.799
AR: 0.782
AR@0.5: 0.937
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_8xb64-210e_coco-384x288-7b8db90e_20220923.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_8xb64-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_res101_8xb64-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.728
AP@0.5: 0.904
AP@0.75: 0.809
AR: 0.783
AR@0.5: 0.942
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_8xb64-210e_coco-256x192-065d3625_20220926.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_8xb32-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_res101_8xb32-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.749
AP@0.5: 0.906
AP@0.75: 0.817
AR: 0.799
AR@0.5: 0.941
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_8xb64-210e_coco-256x192-065d3625_20220926.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_8xb32-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_res152_8xb32-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.736
AP@0.5: 0.904
AP@0.75: 0.818
AR: 0.791
AR@0.5: 0.942
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_8xb32-210e_coco-256x192-0345f330_20220928.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_8xb32-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_res152_8xb32-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.75
AP@0.5: 0.908
AP@0.75: 0.821
AR: 0.8
AR@0.5: 0.942
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_8xb32-210e_coco-384x288-7fbb906f_20220927.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_fp16-8xb64-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_res50_fp16-8xb64-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.716
AP@0.5: 0.898
AP@0.75: 0.798
AR: 0.772
AR@0.5: 0.937
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_fp16-8xb64-210e_coco-256x192-463da051_20220927.pth
<!-- [ALGORITHM] -->
<details>
<summary align="right"><a href="http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html">SimpleBaseline2D (ECCV'2018)</a></summary>
```bibtex
@inproceedings{xiao2018simple,
title={Simple baselines for human pose estimation and tracking},
author={Xiao, Bin and Wu, Haiping and Wei, Yichen},
booktitle={Proceedings of the European conference on computer vision (ECCV)},
pages={466--481},
year={2018}
}
```
</details>
<!-- [BACKBONE] -->
<details>
<summary align="right"><a href="http://openaccess.thecvf.com/content_cvpr_2016/html/He_Deep_Residual_Learning_CVPR_2016_paper.html">ResNet (CVPR'2016)</a></summary>
```bibtex
@inproceedings{he2016deep,
title={Deep residual learning for image recognition},
author={He, Kaiming and Zhang, Xiangyu and Ren, Shaoqing and Sun, Jian},
booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
pages={770--778},
year={2016}
}
```
</details>
<!-- [ALGORITHM] -->
<details>
<summary align="right"><a href="http://openaccess.thecvf.com/content_CVPR_2020/html/Zhang_Distribution-Aware_Coordinate_Representation_for_Human_Pose_Estimation_CVPR_2020_paper.html">DarkPose (CVPR'2020)</a></summary>
```bibtex
@inproceedings{zhang2020distribution,
title={Distribution-aware coordinate representation for human pose estimation},
author={Zhang, Feng and Zhu, Xiatian and Dai, Hanbin and Ye, Mao and Zhu, Ce},
booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
pages={7093--7102},
year={2020}
}
```
</details>
<!-- [DATASET] -->
<details>
<summary align="right"><a href="https://link.springer.com/chapter/10.1007/978-3-319-10602-1_48">COCO (ECCV'2014)</a></summary>
```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}
}
```
</details>
Results on COCO val2017 with detector having human AP of 56.4 on COCO val2017 dataset
| Arch | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [pose_resnet_50_dark](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_dark-8xb64-210e_coco-256x192.py) | 256x192 | 0.724 | 0.897 | 0.797 | 0.777 | 0.934 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_dark-8xb64-210e_coco-256x192-c129dcb6_20220926.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_dark-8xb64-210e_coco-256x192_20220926.log) |
| [pose_resnet_50_dark](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_dark-8xb64-210e_coco-384x288.py) | 384x288 | 0.735 | 0.902 | 0.801 | 0.786 | 0.938 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_dark-8xb64-210e_coco-384x288-8b90b538_20220926.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_dark-8xb64-210e_coco-384x288_20220926.log) |
| [pose_resnet_101_dark](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_dark-8xb64-210e_coco-256x192.py) | 256x192 | 0.733 | 0.900 | 0.810 | 0.786 | 0.938 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_dark-8xb64-210e_coco-256x192-528ec248_20220926.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_dark-8xb64-210e_coco-256x192_20220926.log) |
| [pose_resnet_101_dark](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_dark-8xb64-210e_coco-384x288.py) | 384x288 | 0.749 | 0.905 | 0.818 | 0.799 | 0.940 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_dark-8xb64-210e_coco-384x288-487d40a4_20220926.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_dark-8xb64-210e_coco-384x288_20220926.log) |
| [pose_resnet_152_dark](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_dark-8xb32-210e_coco-256x192.py) | 256x192 | 0.743 | 0.906 | 0.819 | 0.796 | 0.943 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_dark-8xb32-210e_coco-256x192-f754df5f_20221031.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_dark-8xb32-210e_coco-256x192_20221031.log) |
| [pose_resnet_152_dark](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_dark-8xb32-210e_coco-384x288.py) | 384x288 | 0.755 | 0.907 | 0.825 | 0.805 | 0.943 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_dark-8xb32-210e_coco-384x288-329f8454_20221031.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_dark-8xb32-210e_coco-384x288_20221031.log) |
Models:
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_dark-8xb64-210e_coco-256x192.py
In Collection: DarkPose
Metadata:
Architecture: &id001
- SimpleBaseline2D
- ResNet
- DarkPose
Training Data: COCO
Name: td-hm_res50_dark-8xb64-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.724
AP@0.5: 0.897
AP@0.75: 0.797
AR: 0.777
AR@0.5: 0.934
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_dark-8xb64-210e_coco-256x192-c129dcb6_20220926.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_dark-8xb64-210e_coco-384x288.py
In Collection: DarkPose
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_res50_dark-8xb64-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.735
AP@0.5: 0.902
AP@0.75: 0.801
AR: 0.786
AR@0.5: 0.938
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_dark-8xb64-210e_coco-384x288-8b90b538_20220926.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_dark-8xb64-210e_coco-256x192.py
In Collection: DarkPose
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_res101_dark-8xb64-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.733
AP@0.5: 0.9
AP@0.75: 0.81
AR: 0.786
AR@0.5: 0.938
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_dark-8xb64-210e_coco-256x192-528ec248_20220926.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_dark-8xb64-210e_coco-384x288.py
In Collection: DarkPose
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_res101_dark-8xb64-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.749
AP@0.5: 0.905
AP@0.75: 0.818
AR: 0.799
AR@0.5: 0.94
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res101_dark-8xb64-210e_coco-384x288-487d40a4_20220926.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_dark-8xb32-210e_coco-256x192.py
In Collection: DarkPose
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_res152_dark-8xb32-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.743
AP@0.5: 0.906
AP@0.75: 0.819
AR: 0.796
AR@0.5: 0.943
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_dark-8xb32-210e_coco-256x192-f754df5f_20221031.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_dark-8xb32-210e_coco-384x288.py
In Collection: DarkPose
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_res152_dark-8xb32-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.757
AP@0.5: 0.907
AP@0.75: 0.825
AR: 0.805
AR@0.5: 0.943
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res152_dark-8xb32-210e_coco-384x288-329f8454_20221031.pth
<!-- [ALGORITHM] -->
<details>
<summary align="right"><a href="http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html">SimpleBaseline2D (ECCV'2018)</a></summary>
```bibtex
@inproceedings{xiao2018simple,
title={Simple baselines for human pose estimation and tracking},
author={Xiao, Bin and Wu, Haiping and Wei, Yichen},
booktitle={Proceedings of the European conference on computer vision (ECCV)},
pages={466--481},
year={2018}
}
```
</details>
<!-- [BACKBONE] -->
<details>
<summary align="right"><a href="http://openaccess.thecvf.com/content_cvpr_2016/html/He_Deep_Residual_Learning_CVPR_2016_paper.html">ResNet (CVPR'2016)</a></summary>
```bibtex
@inproceedings{he2016deep,
title={Deep residual learning for image recognition},
author={He, Kaiming and Zhang, Xiangyu and Ren, Shaoqing and Sun, Jian},
booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
pages={770--778},
year={2016}
}
```
</details>
<!-- [OTHERS] -->
<details>
<summary align="right"><a href="https://arxiv.org/abs/1710.03740">FP16 (ArXiv'2017)</a></summary>
```bibtex
@article{micikevicius2017mixed,
title={Mixed precision training},
author={Micikevicius, Paulius and Narang, Sharan and Alben, Jonah and Diamos, Gregory and Elsen, Erich and Garcia, David and Ginsburg, Boris and Houston, Michael and Kuchaiev, Oleksii and Venkatesh, Ganesh and others},
journal={arXiv preprint arXiv:1710.03740},
year={2017}
}
```
</details>
<!-- [DATASET] -->
<details>
<summary align="right"><a href="https://link.springer.com/chapter/10.1007/978-3-319-10602-1_48">COCO (ECCV'2014)</a></summary>
```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}
}
```
</details>
Results on COCO val2017 with detector having human AP of 56.4 on COCO val2017 dataset
| Arch | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [pose_resnet_50_fp16](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_fp16-8xb64-210e_coco-256x192.py) | 256x192 | 0.716 | 0.898 | 0.798 | 0.772 | 0.937 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_fp16-8xb64-210e_coco-256x192-463da051_20220927.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_res50_fp16-8xb64-210e_coco-256x192_20220927.log) |
<!-- [BACKBONE] -->
<details>
<summary align="right"><a href="http://openaccess.thecvf.com/content_CVPR_2019/html/He_Bag_of_Tricks_for_Image_Classification_with_Convolutional_Neural_Networks_CVPR_2019_paper.html">ResNetV1D (CVPR'2019)</a></summary>
```bibtex
@inproceedings{he2019bag,
title={Bag of tricks for image classification with convolutional neural networks},
author={He, Tong and Zhang, Zhi and Zhang, Hang and Zhang, Zhongyue and Xie, Junyuan and Li, Mu},
booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
pages={558--567},
year={2019}
}
```
</details>
<!-- [DATASET] -->
<details>
<summary align="right"><a href="https://link.springer.com/chapter/10.1007/978-3-319-10602-1_48">COCO (ECCV'2014)</a></summary>
```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}
}
```
</details>
Results on COCO val2017 with detector having human AP of 56.4 on COCO val2017 dataset
| Arch | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [pose_resnetv1d_50](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d50_8xb64-210e_coco-256x192.py) | 256x192 | 0.722 | 0.897 | 0.796 | 0.777 | 0.936 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d50_8xb64-210e_coco-256x192-27545d63_20221020.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d50_8xb64-210e_coco-256x192_20221020.log) |
| [pose_resnetv1d_50](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d50_8xb64-210e_coco-384x288.py) | 384x288 | 0.730 | 0.899 | 0.800 | 0.782 | 0.935 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d50_8xb64-210e_coco-384x288-0646b46e_20221020.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d50_8xb64-210e_coco-384x288_20221020.log) |
| [pose_resnetv1d_101](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d101_8xb64-210e_coco-256x192.py) | 256x192 | 0.732 | 0.901 | 0.808 | 0.785 | 0.940 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d101_8xb64-210e_coco-256x192-ee9e7212_20221021.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d101_8xb64-210e_coco-256x192_20221021.log) |
| [pose_resnetv1d_101](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d101_8xb32-210e_coco-384x288.py) | 384x288 | 0.748 | 0.906 | 0.817 | 0.798 | 0.941 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d101_8xb32-210e_coco-384x288-d0b5875f_20221028.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d101_8xb32-210e_coco-384x288_20221028.log) |
| [pose_resnetv1d_152](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d152_8xb32-210e_coco-256x192.py) | 256x192 | 0.737 | 0.904 | 0.814 | 0.790 | 0.940 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d152_8xb32-210e_coco-256x192-fd49f947_20221021.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d152_8xb32-210e_coco-256x192_20221021.log) |
| [pose_resnetv1d_152](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d152_8xb48-210e_coco-384x288.py) | 384x288 | 0.751 | 0.907 | 0.821 | 0.801 | 0.942 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d152_8xb48-210e_coco-384x288-b9a99602_20221022.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d152_8xb48-210e_coco-384x288_20221022.log) |
Models:
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d50_8xb64-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: &id001
- SimpleBaseline2D
- ResNetV1D
Training Data: COCO
Name: td-hm_resnetv1d50_8xb64-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.722
AP@0.5: 0.897
AP@0.75: 0.796
AR: 0.777
AR@0.5: 0.936
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d50_8xb64-210e_coco-256x192-27545d63_20221020.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d50_8xb64-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_resnetv1d50_8xb64-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.73
AP@0.5: 0.899
AP@0.75: 0.8
AR: 0.782
AR@0.5: 0.935
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d50_8xb64-210e_coco-384x288-0646b46e_20221020.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d101_8xb64-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_resnetv1d101_8xb64-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.732
AP@0.5: 0.901
AP@0.75: 0.808
AR: 0.785
AR@0.5: 0.940
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d101_8xb64-210e_coco-256x192-ee9e7212_20221021.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d101_8xb32-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_resnetv1d101_8xb32-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.748
AP@0.5: 0.906
AP@0.75: 0.817
AR: 0.798
AR@0.5: 0.941
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d101_8xb32-210e_coco-384x288-d0b5875f_20221028.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d152_8xb32-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_resnetv1d152_8xb32-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.737
AP@0.5: 0.904
AP@0.75: 0.814
AR: 0.790
AR@0.5: 0.94
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d152_8xb32-210e_coco-256x192-fd49f947_20221021.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d152_8xb48-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_resnetv1d152_8xb48-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.751
AP@0.5: 0.907
AP@0.75: 0.821
AR: 0.801
AR@0.5: 0.942
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnetv1d152_8xb48-210e_coco-384x288-b9a99602_20221022.pth
<!-- [BACKBONE] -->
<details>
<summary align="right"><a href="http://openaccess.thecvf.com/content_cvpr_2017/html/Xie_Aggregated_Residual_Transformations_CVPR_2017_paper.html">ResNext (CVPR'2017)</a></summary>
```bibtex
@inproceedings{xie2017aggregated,
title={Aggregated residual transformations for deep neural networks},
author={Xie, Saining and Girshick, Ross and Doll{\'a}r, Piotr and Tu, Zhuowen and He, Kaiming},
booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
pages={1492--1500},
year={2017}
}
```
</details>
<!-- [DATASET] -->
<details>
<summary align="right"><a href="https://link.springer.com/chapter/10.1007/978-3-319-10602-1_48">COCO (ECCV'2014)</a></summary>
```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}
}
```
</details>
Results on COCO val2017 with detector having human AP of 56.4 on COCO val2017 dataset
| Arch | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [pose_resnext_50](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnext50_8xb64-210e_coco-256x192.py) | 256x192 | 0.715 | 0.897 | 0.791 | 0.771 | 0.935 | [ckpt](https://download.openmmlab.com/mmpose/top_down/resnext/resnext50_coco_256x192-dcff15f6_20200727.pth) | [log](https://download.openmmlab.com/mmpose/top_down/resnext/resnext50_coco_256x192_20200727.log.json) |
| [pose_resnext_50](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnext50_8xb64-210e_coco-384x288.py) | 384x288 | 0.724 | 0.899 | 0.794 | 0.777 | 0.936 | [ckpt](https://download.openmmlab.com/mmpose/top_down/resnext/resnext50_coco_384x288-412c848f_20200727.pth) | [log](https://download.openmmlab.com/mmpose/top_down/resnext/resnext50_coco_384x288_20200727.log.json) |
| [pose_resnext_101](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnext101_8xb64-210e_coco-256x192.py) | 256x192 | 0.726 | 0.900 | 0.801 | 0.781 | 0.939 | [ckpt](https://download.openmmlab.com/mmpose/top_down/resnext/resnext101_coco_256x192-c7eba365_20200727.pth) | [log](https://download.openmmlab.com/mmpose/top_down/resnext/resnext101_coco_256x192_20200727.log.json) |
| [pose_resnext_101](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnext101_8xb32-210e_coco-384x288.py) | 384x288 | 0.744 | 0.903 | 0.815 | 0.794 | 0.939 | [ckpt](https://download.openmmlab.com/mmpose/top_down/resnext/resnext101_coco_384x288-f5eabcd6_20200727.pth) | [log](https://download.openmmlab.com/mmpose/top_down/resnext/resnext101_coco_384x288_20200727.log.json) |
| [pose_resnext_152](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnext152_8xb32-210e_coco-256x192.py) | 256x192 | 0.730 | 0.903 | 0.808 | 0.785 | 0.940 | [ckpt](https://download.openmmlab.com/mmpose/top_down/resnext/resnext152_coco_256x192-102449aa_20200727.pth) | [log](https://download.openmmlab.com/mmpose/top_down/resnext/resnext152_coco_256x192_20200727.log.json) |
| [pose_resnext_152](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnext152_8xb48-210e_coco-384x288.py) | 384x288 | 0.742 | 0.904 | 0.810 | 0.794 | 0.940 | [ckpt](https://download.openmmlab.com/mmpose/top_down/resnext/resnext152_coco_384x288-806176df_20200727.pth) | [log](https://download.openmmlab.com/mmpose/top_down/resnext/resnext152_coco_384x288_20200727.log.json) |
Models:
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnext50_8xb64-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: &id001
- SimpleBaseline2D
- ResNext
Training Data: COCO
Name: td-hm_resnext50_8xb64-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.715
AP@0.5: 0.897
AP@0.75: 0.791
AR: 0.771
AR@0.5: 0.935
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/resnext/resnext50_coco_256x192-dcff15f6_20200727.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnext50_8xb64-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_resnext50_8xb64-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.724
AP@0.5: 0.899
AP@0.75: 0.794
AR: 0.777
AR@0.5: 0.936
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/resnext/resnext50_coco_384x288-412c848f_20200727.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnext101_8xb64-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_resnext101_8xb64-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.726
AP@0.5: 0.9
AP@0.75: 0.801
AR: 0.781
AR@0.5: 0.939
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/resnext/resnext101_coco_256x192-c7eba365_20200727.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnext101_8xb32-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_resnext101_8xb32-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.744
AP@0.5: 0.903
AP@0.75: 0.815
AR: 0.794
AR@0.5: 0.939
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/resnext/resnext101_coco_384x288-f5eabcd6_20200727.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnext152_8xb32-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_resnext152_8xb32-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.73
AP@0.5: 0.903
AP@0.75: 0.808
AR: 0.785
AR@0.5: 0.94
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/resnext/resnext152_coco_256x192-102449aa_20200727.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_resnext152_8xb48-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_resnext152_8xb48-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.742
AP@0.5: 0.904
AP@0.75: 0.81
AR: 0.794
AR@0.5: 0.94
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/resnext/resnext152_coco_384x288-806176df_20200727.pth
<!-- [ALGORITHM] -->
<details>
<summary align="right"><a href="https://link.springer.com/chapter/10.1007/978-3-030-58580-8_27">RSN (ECCV'2020)</a></summary>
```bibtex
@misc{cai2020learning,
title={Learning Delicate Local Representations for Multi-Person Pose Estimation},
author={Yuanhao Cai and Zhicheng Wang and Zhengxiong Luo and Binyi Yin and Angang Du and Haoqian Wang and Xinyu Zhou and Erjin Zhou and Xiangyu Zhang and Jian Sun},
year={2020},
eprint={2003.04030},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
</details>
<!-- [DATASET] -->
<details>
<summary align="right"><a href="https://link.springer.com/chapter/10.1007/978-3-319-10602-1_48">COCO (ECCV'2014)</a></summary>
```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}
}
```
</details>
Results on COCO val2017 with detector having human AP of 56.4 on COCO val2017 dataset
| Arch | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [rsn_18](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_rsn18_8xb32-210e_coco-256x192.py) | 256x192 | 0.704 | 0.887 | 0.781 | 0.773 | 0.927 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_rsn18_8xb32-210e_coco-256x192-9049ed09_20221013.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_rsn18_8xb32-210e_coco-256x192_20221013.log) |
| [rsn_50](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_rsn50_8xb32-210e_coco-256x192.py) | 256x192 | 0.724 | 0.894 | 0.799 | 0.790 | 0.935 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_rsn50_8xb32-210e_coco-256x192-c35901d5_20221013.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_rsn50_8xb32-210e_coco-256x192_20221013.log) |
| [2xrsn_50](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_2xrsn50_8xb32-210e_coco-256x192.py) | 256x192 | 0.748 | 0.900 | 0.821 | 0.810 | 0.939 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_2xrsn50_8xb32-210e_coco-256x192-9ede341e_20221013.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_2xrsn50_8xb32-210e_coco-256x192_20221013.log) |
| [3xrsn_50](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_3xrsn50_8xb32-210e_coco-256x192.py) | 256x192 | 0.750 | 0.900 | 0.824 | 0.814 | 0.941 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_3xrsn50_8xb32-210e_coco-256x192-c3e3c4fe_20221013.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_3xrsn50_8xb32-210e_coco-256x192_20221013.log) |
Collections:
- Name: RSN
Paper:
Title: Learning Delicate Local Representations for Multi-Person Pose Estimation
URL: https://link.springer.com/chapter/10.1007/978-3-030-58580-8_27
README: https://github.com/open-mmlab/mmpose/blob/main/docs/src/papers/backbones/rsn.md
Models:
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_rsn18_8xb32-210e_coco-256x192.py
In Collection: RSN
Metadata:
Architecture: &id001
- RSN
Training Data: COCO
Name: td-hm_rsn18_8xb32-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.704
AP@0.5: 0.887
AP@0.75: 0.781
AR: 0.773
AR@0.5: 0.927
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_rsn18_8xb32-210e_coco-256x192-9049ed09_20221013.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_rsn50_8xb32-210e_coco-256x192.py
In Collection: RSN
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_rsn50_8xb32-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.724
AP@0.5: 0.894
AP@0.75: 0.799
AR: 0.79
AR@0.5: 0.935
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_rsn50_8xb32-210e_coco-256x192-c35901d5_20221013.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_2xrsn50_8xb32-210e_coco-256x192.py
In Collection: RSN
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_2xrsn50_8xb32-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.748
AP@0.5: 0.9
AP@0.75: 0.821
AR: 0.81
AR@0.5: 0.939
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_2xrsn50_8xb32-210e_coco-256x192-9ede341e_20221013.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_3xrsn50_8xb32-210e_coco-256x192.py
In Collection: RSN
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_3xrsn50_8xb32-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.75
AP@0.5: 0.9
AP@0.75: 0.824
AR: 0.814
AR@0.5: 0.941
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_3xrsn50_8xb32-210e_coco-256x192-c3e3c4fe_20221013.pth
<!-- [ALGORITHM] -->
<details>
<summary align="right"><a href="http://openaccess.thecvf.com/content_CVPR_2020/html/Liu_Improving_Convolutional_Networks_With_Self-Calibrated_Convolutions_CVPR_2020_paper.html">SCNet (CVPR'2020)</a></summary>
```bibtex
@inproceedings{liu2020improving,
title={Improving Convolutional Networks with Self-Calibrated Convolutions},
author={Liu, Jiang-Jiang and Hou, Qibin and Cheng, Ming-Ming and Wang, Changhu and Feng, Jiashi},
booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
pages={10096--10105},
year={2020}
}
```
</details>
<!-- [DATASET] -->
<details>
<summary align="right"><a href="https://link.springer.com/chapter/10.1007/978-3-319-10602-1_48">COCO (ECCV'2014)</a></summary>
```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}
}
```
</details>
Results on COCO val2017 with detector having human AP of 56.4 on COCO val2017 dataset
| Arch | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [pose_scnet_50](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_scnet50_8xb64-210e_coco-256x192.py) | 256x192 | 0.728 | 0.899 | 0.807 | 0.784 | 0.938 | [ckpt](https://download.openmmlab.com/mmpose/top_down/scnet/scnet50_coco_256x192-6920f829_20200709.pth) | [log](https://download.openmmlab.com/mmpose/top_down/scnet/scnet50_coco_256x192_20200709.log.json) |
| [pose_scnet_50](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_scnet50_8xb32-210e_coco-384x288.py) | 384x288 | 0.751 | 0.906 | 0.818 | 0.802 | 0.942 | [ckpt](https://download.openmmlab.com/mmpose/top_down/scnet/scnet50_coco_384x288-9cacd0ea_20200709.pth) | [log](https://download.openmmlab.com/mmpose/top_down/scnet/scnet50_coco_384x288_20200709.log.json) |
| [pose_scnet_101](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_scnet101_8xb32-210e_coco-256x192.py) | 256x192 | 0.733 | 0.902 | 0.811 | 0.789 | 0.940 | [ckpt](https://download.openmmlab.com/mmpose/top_down/scnet/scnet101_coco_256x192-6d348ef9_20200709.pth) | [log](https://download.openmmlab.com/mmpose/top_down/scnet/scnet101_coco_256x192_20200709.log.json) |
| [pose_scnet_101](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_scnet101_8xb48-210e_coco-384x288.py) | 384x288 | 0.752 | 0.906 | 0.823 | 0.804 | 0.943 | [ckpt](https://download.openmmlab.com/mmpose/top_down/scnet/scnet101_coco_384x288-0b6e631b_20200709.pth) | [log](https://download.openmmlab.com/mmpose/top_down/scnet/scnet101_coco_384x288_20200709.log.json) |
Models:
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_scnet50_8xb64-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: &id001
- SCNet
Training Data: COCO
Name: td-hm_scnet50_8xb64-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.728
AP@0.5: 0.899
AP@0.75: 0.807
AR: 0.784
AR@0.5: 0.938
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/scnet/scnet50_coco_256x192-6920f829_20200709.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_scnet50_8xb32-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: topdown_heatmap_scnet50_coco_384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.751
AP@0.5: 0.906
AP@0.75: 0.818
AR: 0.802
AR@0.5: 0.942
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/scnet/scnet50_coco_384x288-9cacd0ea_20200709.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_scnet101_8xb32-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_scnet101_8xb32-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.733
AP@0.5: 0.902
AP@0.75: 0.811
AR: 0.789
AR@0.5: 0.94
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/scnet/scnet101_coco_256x192-6d348ef9_20200709.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_scnet101_8xb48-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_scnet101_8xb48-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.752
AP@0.5: 0.906
AP@0.75: 0.823
AR: 0.804
AR@0.5: 0.943
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/scnet/scnet101_coco_384x288-0b6e631b_20200709.pth
<!-- [BACKBONE] -->
<details>
<summary align="right"><a href="http://openaccess.thecvf.com/content_cvpr_2018/html/Hu_Squeeze-and-Excitation_Networks_CVPR_2018_paper">SEResNet (CVPR'2018)</a></summary>
```bibtex
@inproceedings{hu2018squeeze,
title={Squeeze-and-excitation networks},
author={Hu, Jie and Shen, Li and Sun, Gang},
booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
pages={7132--7141},
year={2018}
}
```
</details>
<!-- [DATASET] -->
<details>
<summary align="right"><a href="https://link.springer.com/chapter/10.1007/978-3-319-10602-1_48">COCO (ECCV'2014)</a></summary>
```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}
}
```
</details>
Results on COCO val2017 with detector having human AP of 56.4 on COCO val2017 dataset
| Arch | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [pose_seresnet_50](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_seresnet50_8xb64-210e_coco-256x192.py) | 256x192 | 0.729 | 0.903 | 0.807 | 0.784 | 0.941 | [ckpt](https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet50_coco_256x192-25058b66_20200727.pth) | [log](https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet50_coco_256x192_20200727.log.json) |
| [pose_seresnet_50](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_seresnet50_8xb64-210e_coco-384x288.py) | 384x288 | 0.748 | 0.904 | 0.819 | 0.799 | 0.941 | [ckpt](https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet50_coco_384x288-bc0b7680_20200727.pth) | [log](https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet50_coco_384x288_20200727.log.json) |
| [pose_seresnet_101](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_seresnet101_8xb64-210e_coco-256x192.py) | 256x192 | 0.734 | 0.905 | 0.814 | 0.790 | 0.941 | [ckpt](https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet101_coco_256x192-83f29c4d_20200727.pth) | [log](https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet101_coco_256x192_20200727.log.json) |
| [pose_seresnet_101](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_seresnet101_8xb32-210e_coco-384x288.py) | 384x288 | 0.754 | 0.907 | 0.823 | 0.805 | 0.943 | [ckpt](https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet101_coco_384x288-48de1709_20200727.pth) | [log](https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet101_coco_384x288_20200727.log.json) |
| [pose_seresnet_152\*](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_seresnet152_8xb32-210e_coco-256x192.py) | 256x192 | 0.730 | 0.899 | 0.810 | 0.787 | 0.939 | [ckpt](https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet152_coco_256x192-1c628d79_20200727.pth) | [log](https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet152_coco_256x192_20200727.log.json) |
| [pose_seresnet_152\*](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_seresnet152_8xb48-210e_coco-384x288.py) | 384x288 | 0.753 | 0.906 | 0.824 | 0.806 | 0.945 | [ckpt](https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet152_coco_384x288-58b23ee8_20200727.pth) | [log](https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet152_coco_384x288_20200727.log.json) |
Note that * means without imagenet pre-training.
Models:
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_seresnet50_8xb64-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: &id001
- SEResNet
Training Data: COCO
Name: td-hm_seresnet50_8xb64-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.729
AP@0.5: 0.903
AP@0.75: 0.807
AR: 0.784
AR@0.5: 0.941
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet50_coco_256x192-25058b66_20200727.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_seresnet50_8xb64-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_seresnet50_8xb64-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.748
AP@0.5: 0.904
AP@0.75: 0.819
AR: 0.799
AR@0.5: 0.941
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet50_coco_384x288-bc0b7680_20200727.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_seresnet101_8xb64-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_seresnet101_8xb64-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.734
AP@0.5: 0.905
AP@0.75: 0.814
AR: 0.79
AR@0.5: 0.941
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet101_coco_256x192-83f29c4d_20200727.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_seresnet101_8xb32-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_seresnet101_8xb32-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.754
AP@0.5: 0.907
AP@0.75: 0.823
AR: 0.805
AR@0.5: 0.943
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet101_coco_384x288-48de1709_20200727.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_seresnet152_8xb32-210e_coco-256x192.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_seresnet152_8xb32-210e_coco-256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.73
AP@0.5: 0.899
AP@0.75: 0.81
AR: 0.787
AR@0.5: 0.939
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet152_coco_256x192-1c628d79_20200727.pth
- Config: configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_seresnet152_8xb48-210e_coco-384x288.py
In Collection: SimpleBaseline2D
Metadata:
Architecture: *id001
Training Data: COCO
Name: td-hm_seresnet152_8xb48-210e_coco-384x288
Results:
- Dataset: COCO
Metrics:
AP: 0.753
AP@0.5: 0.906
AP@0.75: 0.824
AR: 0.806
AR@0.5: 0.945
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/seresnet/seresnet152_coco_384x288-58b23ee8_20200727.pth
<!-- [BACKBONE] -->
<details>
<summary align="right"><a href="http://openaccess.thecvf.com/content_cvpr_2018/html/Zhang_ShuffleNet_An_Extremely_CVPR_2018_paper.html">ShufflenetV1 (CVPR'2018)</a></summary>
```bibtex
@inproceedings{zhang2018shufflenet,
title={Shufflenet: An extremely efficient convolutional neural network for mobile devices},
author={Zhang, Xiangyu and Zhou, Xinyu and Lin, Mengxiao and Sun, Jian},
booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
pages={6848--6856},
year={2018}
}
```
</details>
<!-- [DATASET] -->
<details>
<summary align="right"><a href="https://link.springer.com/chapter/10.1007/978-3-319-10602-1_48">COCO (ECCV'2014)</a></summary>
```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}
}
```
</details>
Results on COCO val2017 with detector having human AP of 56.4 on COCO val2017 dataset
| Arch | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [pose_shufflenetv1](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_shufflenetv1_8xb64-210e_coco-256x192.py) | 256x192 | 0.587 | 0.849 | 0.654 | 0.654 | 0.896 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_shufflenetv1_8xb64-210e_coco-256x192-7a7ea4f4_20221013.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_shufflenetv1_8xb64-210e_coco-256x192_20221013.log) |
| [pose_shufflenetv1](/configs/body_2d_keypoint/topdown_heatmap/coco/td-hm_shufflenetv1_8xb64-210e_coco-384x288.py) | 384x288 | 0.626 | 0.862 | 0.696 | 0.687 | 0.903 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_shufflenetv1_8xb64-210e_coco-384x288-8342f8ba_20221013.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/coco/td-hm_shufflenetv1_8xb64-210e_coco-384x288_20221013.log) |
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment