# STGCN
## Introduction
STGCN is one of the first algorithms that adopt Graph Convolution Neural Networks for skeleton processing. We provide STGCN trained on NTURGB+D with 2D skeletons (HRNet) and 3D skeletons in both the original training setting and the **PYSKL** training setting. We provide checkpoints for four modalities: Joint, Bone, Joint Motion, and Bone Motion. The accuracy of each modality links to the weight file.
## Citation
```BibTeX
@inproceedings{yan2018spatial,
title={Spatial temporal graph convolutional networks for skeleton-based action recognition},
author={Yan, Sijie and Xiong, Yuanjun and Lin, Dahua},
booktitle={Thirty-second AAAI conference on artificial intelligence},
year={2018}
}
# If you use the STGCN with PYSKL practices in your work
@misc{duan2022pyskl,
title={PYSKL: a toolbox for skeleton-based video understanding},
author={PYSKL Contributors},
howpublished = {\url{https://github.com/kennymckormick/pyskl}},
year={2022}
}
```
## Model Zoo
We release numerous checkpoints trained with various modalities, annotations on NTURGB+D and NTURGB+D 120. The accuracy of each modality links to the weight file.
| Dataset | Practice | Annotation | GPUs | Training Epochs | Joint Top1<br/>Config Link: Weight Link | Bone Top1<br/>Config Link: Weight Link | Joint Motion Top1<br/>Config Link: Weight Link | Bone-Motion Top1<br/>Config Link: Weight Link | Two-Stream Top1 | Four Stream Top1 |
| :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: |
| NTURGB+D XSub | Vanilla | Official 3D Skeleton | 8 | 80 | [joint_config](/configs/stgcn/stgcn_vanilla_ntu60_xsub_3dkp/j.py): [81.5](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_vanilla_ntu60_xsub_3dkp/j.pth) | [bone_config](/configs/stgcn/stgcn_vanilla_ntu60_xsub_3dkp/b.py): [81.0](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_vanilla_ntu60_xsub_3dkp/b.pth) | [joint_motion_config](/configs/stgcn/stgcn_vanilla_ntu60_xsub_3dkp/jm.py): [79.9](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_vanilla_ntu60_xsub_3dkp/jm.pth) | [bone_motion_config](/configs/stgcn/stgcn_vanilla_ntu60_xsub_3dkp/bm.py): [81.2](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_vanilla_ntu60_xsub_3dkp/bm.pth) | 84.3 | 86.6 |
| NTURGB+D XSub | Vanilla | HRNet 2D Skeleton | 8 | 80 | [joint_config](/configs/stgcn/stgcn_vanilla_ntu60_xsub_hrnet/j.py): [85.7](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_vanilla_ntu60_xsub_hrnet/j.pth) | [bone_config](/configs/stgcn/stgcn_vanilla_ntu60_xsub_hrnet/b.py): [85.8](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_vanilla_ntu60_xsub_hrnet/b.pth) | [joint_motion_config](/configs/stgcn/stgcn_vanilla_ntu60_xsub_hrnet/jm.py): [81.6](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_vanilla_ntu60_xsub_hrnet/jm.pth) | [bone_motion_config](/configs/stgcn/stgcn_vanilla_ntu60_xsub_hrnet/bm.py): [83.9](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_vanilla_ntu60_xsub_hrnet/bm.pth) | 88.8 | 90.1 |
| NTURGB+D XSub | PYSKL | Official 3D Skeleton | 8 | 80 | [joint_config](/configs/stgcn/stgcn_pyskl_ntu60_xsub_3dkp/j.py): [87.8](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_pyskl_ntu60_xsub_3dkp/j.pth) | [bone_config](/configs/stgcn/stgcn_pyskl_ntu60_xsub_3dkp/b.py): [88.6](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_pyskl_ntu60_xsub_3dkp/b.pth) | [joint_motion_config](/configs/stgcn/stgcn_pyskl_ntu60_xsub_3dkp/jm.py): [85.8](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_pyskl_ntu60_xsub_3dkp/jm.pth) | [bone_motion_config](/configs/stgcn/stgcn_pyskl_ntu60_xsub_3dkp/bm.py): [86.2](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_pyskl_ntu60_xsub_3dkp/bm.pth) | 90.0 | 90.7 |
| NTURGB+D XSub | PYSKL | HRNet 2D Skeleton | 8 | 80 | [joint_config](/configs/stgcn/stgcn_pyskl_ntu60_xsub_hrnet/j.py): [89.0](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_pyskl_ntu60_xsub_hrnet/j.pth) | [bone_config](/configs/stgcn/stgcn_pyskl_ntu60_xsub_hrnet/b.py): [91.2](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_pyskl_ntu60_xsub_hrnet/b.pth) | [joint_motion_config](/configs/stgcn/stgcn_pyskl_ntu60_xsub_hrnet/jm.py): [86.7](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_pyskl_ntu60_xsub_hrnet/jm.pth) | [bone_motion_config](/configs/stgcn/stgcn_pyskl_ntu60_xsub_hrnet/bm.py): [87.8](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_pyskl_ntu60_xsub_hrnet/bm.pth) | 92.0 | 92.4 |
| NTURGB+D XView | Vanilla | Official 3D Skeleton | 8 | 80 | [joint_config](/configs/stgcn/stgcn_vanilla_ntu60_xview_3dkp/j.py): [90.1](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_vanilla_ntu60_xview_3dkp/j.pth) | [bone_config](/configs/stgcn/stgcn_vanilla_ntu60_xview_3dkp/b.py): [87.7](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_vanilla_ntu60_xview_3dkp/b.pth) | [joint_motion_config](/configs/stgcn/stgcn_vanilla_ntu60_xview_3dkp/jm.py): [88.8](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_vanilla_ntu60_xview_3dkp/jm.pth) | [bone_motion_config](/configs/stgcn/stgcn_vanilla_ntu60_xview_3dkp/bm.py): [88.3](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_vanilla_ntu60_xview_3dkp/bm.pth) | 91.4 | 93.2 |
| NTURGB+D XView | Vanilla | HRNet 2D Skeleton | 8 | 80 | [joint_config](/configs/stgcn/stgcn_vanilla_ntu60_xview_hrnet/j.py): [92.4](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_vanilla_ntu60_xview_hrnet/j.pth) | [bone_config](/configs/stgcn/stgcn_vanilla_ntu60_xview_hrnet/b.py): [90.0](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_vanilla_ntu60_xview_hrnet/b.pth) | [joint_motion_config](/configs/stgcn/stgcn_vanilla_ntu60_xview_hrnet/jm.py): [92.0](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_vanilla_ntu60_xview_hrnet/jm.pth) | [bone_motion_config](/configs/stgcn/stgcn_vanilla_ntu60_xview_hrnet/bm.py): [86.5](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_vanilla_ntu60_xview_hrnet/bm.pth) | 93.8 | 95.1 |
| NTURGB+D XView | PYSKL | Official 3D Skeleton | 8 | 80 | [joint_config](/configs/stgcn/stgcn_pyskl_ntu60_xview_3dkp/j.py): [95.5](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_pyskl_ntu60_xview_3dkp/j.pth) | [bone_config](/configs/stgcn/stgcn_pyskl_ntu60_xview_3dkp/b.py): [95.0](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_pyskl_ntu60_xview_3dkp/b.pth) | [joint_motion_config](/configs/stgcn/stgcn_pyskl_ntu60_xview_3dkp/jm.py): [93.7](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_pyskl_ntu60_xview_3dkp/jm.pth) | [bone_motion_config](/configs/stgcn/stgcn_pyskl_ntu60_xview_3dkp/bm.py): [92.8](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_pyskl_ntu60_xview_3dkp/bm.pth) | 96.2 | 96.5 |
| NTURGB+D XView | PYSKL | HRNet 2D Skeleton | 8 | 80 | [joint_config](/configs/stgcn/stgcn_pyskl_ntu60_xview_hrnet/j.py): [98.0](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_pyskl_ntu60_xview_hrnet/j.pth) | [bone_config](/configs/stgcn/stgcn_pyskl_ntu60_xview_hrnet/b.py): [96.5](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_pyskl_ntu60_xview_hrnet/b.pth) | [joint_motion_config](/configs/stgcn/stgcn_pyskl_ntu60_xview_hrnet/jm.py): [95.6](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_pyskl_ntu60_xview_hrnet/jm.pth) | [bone_motion_config](/configs/stgcn/stgcn_pyskl_ntu60_xview_hrnet/bm.py): [95.4](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_pyskl_ntu60_xview_hrnet/bm.pth) | 98.2 | 98.3 |
| NTURGB+D 120 XSub | PYSKL | Official 3D Skeleton | 8 | 80 | [joint_config](/configs/stgcn/stgcn_pyskl_ntu120_xsub_3dkp/j.py): [82.1](http://download.openmmlab.com/mmaction/pyskl/ckpt/stgcn/stgcn_pyskl_ntu120_xsub_3dkp/j.pth) | [bone_config](/configs/stgcn/stgcn_pyskl_ntu120_xsub_3dkp/b.py): [83.7](http://download.openmmlab.com/mmaction/pyskl/
cpu环境下可运行的骨骼序列行为识别的代码 （1170个子文件） 
ntu_sample.avi 1.07MB
setup.cfg 559B .gitignore 2KB .gitignore 184B pyskl-main.iml 596B visualize_heatmap_volume.ipynb 13KB merge_pretrain.ipynb 6KB inference_speed.ipynb 6KB diving48_example.ipynb 5KB vis_skeleton.ipynb 3KB 20231113_125420.log.json 552KB 20231112_172959.log.json 86KB 20231130_135750.log.json 27KB 20231114_125100.log.json 22KB 20231112_125753.log.json 17KB 20231115_174026.log.json 9KB 20231128_205848.log.json 9KB 20231129_171513.log.json 8KB 20231114_213130.log.json 8KB 20231112_124909.log.json 5KB 20231112_123507.log.json 2KB 20231113_121508.log.json 2KB 20231111_155521.log.json 2KB 20231112_201244.log.json 2KB LICENSE 11KB 20231113_125420.log 633KB 20231112_172959.log 106KB 20231130_135750.log 32KB 20231112_125753.log 27KB 20231114_125100.log 26KB 20231112_124909.log 13KB 20231115_174026.log 12KB 20231128_205848.log 11KB 20231112_123507.log 11KB 20231129_171513.log 10KB 20231113_121508.log 10KB 20231114_213130.log 10KB 20231111_155521.log 10KB 20231112_201244.log 10KB 20231109_171158.log 8KB 20231109_213205.log 8KB 20231111_154151.log 8KB 20231111_153604.log 8KB 20231109_171033.log 8KB 20231111_155307.log 8KB 20231109_171328.log 8KB 20231109_170527.log 8KB 20231109_171221.log 8KB 20231111_155332.log 8KB 20231111_155419.log 8KB 20231114_125027.log 8KB 20231114_152916.log 8KB 20231109_170245.log 7KB 20231112_203326.log 7KB 20231112_203350.log 7KB 20231109_165256.log 7KB 20231115_170101.log 0B 20231115_165852.log 0B README.md 12KB README.md 9KB README.md 9KB README.md 9KB README.md 9KB README.md 8KB README.md 8KB README.md 7KB README.md 7KB README.md 6KB README.md 4KB demo.md 3KB
nofi01012.mp4 1.6MB nofi01512.mp4 1.59MB nofi01412.mp4 1.35MB nofi01112.mp4 1.27MB fi1441.mp4 1.25MB fi0921.mp4 1.2MB fi0971.mp4 1.17MB fi0981.mp4 1.17MB fi0911.mp4 1.15MB fi0961.mp4 1.11MB fi0561.mp4 1.09MB fi0571.mp4 962KB fi0901.mp4 930KB nofi10712.mp4 911KB fi1401.mp4 879KB fi0581.mp4 871KB nofi10612.mp4 860KB nofi06712.mp4 859KB nofi06212.mp4 851KB nofi06612.mp4 849KB nofi06412.mp4 847KB nofi06812.mp4 846KB nofi06512.mp4 843KB nofi06312.mp4 837KB fi1411.mp4 834KB nofi03812.mp4 808KB nofi05312.mp4 734KB nofi05412.mp4 727KB nofi07312.mp4 684KB nofi07012.mp4 682KB共 1170 条
- 1
- 2
- 3
- 4
- 5
- 6
- 12
资源评论
辛勤的程序猿
- 粉丝: 1308
- 资源: 9
最新资源
- 基于BS模式的冷链物流系统pf-springboot毕业项目，适合计算机毕-设、实训项目、大作业学习.zip
- 基于simulink实现的110kV继电保护设计实现
- 这个程序是一个简单的git代理切换器，它的主要功能是帮助用户在需要代理的网络环境和不需要代理的网络环境之间快速切换
- java javacv jt1078 rtp流转flv
- iClient for Classic加载wmts山东天地图示例demo
- lvgl-v7版本代码示例
- 基于模型预测控制的无人驾驶车辆轨迹跟踪问题研究，matlab 附有MATLAB程序与详细的建模过程，研究车辆转向的同学可以作为参考
- 基于SpringBoot+vue的名城小区物业管理系统(含报告).zip
- Xmanager Power Suite 8.0.0005
- 阅后即焚平台系统PHP源码
- C#读写西门子PLC.OPC.数据库.Socket 1、PLC数据通信读写； 2、联合OPC； 3、联合Socket； 4、联合数据库；
- 德普微一级代理 DP023N10TGN TOLL DPMOS N-MOSFET 100V 300A 1.75mΩ
- 在线教育网校系统社区版+带安装说明
- 基于Spring+vue的智慧生活商城系统.zip
- 高可用k8s集群离线部署（五）
- 基于Java的考试系统代码pf-springboot毕业项目，适合计算机毕-设、实训项目、大作业学习.rar
资源上传下载、课程学习等过程中有任何疑问或建议，欢迎提出宝贵意见哦~我们会及时处理！
点击此处反馈
