摘要
This paper presents an action recognition method based on 2D human body node data in video. This method uses the pose estimation algorithm to detect the human body node data in each frame of video information. We get the two-dimensional coordinates and confidence data of the nodes, and optimize the arrangement of these data into a 3D array form similar to images. Finally, we use the classical two-dimensional convolutional neural network to carry out classification training. The test on UCF-101 data set shows that this method can indeed improve the accuracy of action recognition based on RGB information to a certain extent, and reduce the training cost.
| 源语言 | 英语 |
|---|---|
| 主期刊名 | Proceedings of the 40th Chinese Control Conference, CCC 2021 |
| 编辑 | Chen Peng, Jian Sun |
| 出版商 | IEEE Computer Society |
| 页 | 7366-7370 |
| 页数 | 5 |
| ISBN（电子版） | 9789881563804 |
| DOI | |
| 出版状态 | 已出版 - 26 7月 2021 |
| 活动 | 40th Chinese Control Conference, CCC 2021 - Shanghai, 中国 期限: 26 7月 2021 → 28 7月 2021 |
出版系列
| 姓名 | Chinese Control Conference, CCC |
|---|---|
| 卷 | 2021-July |
| ISSN（印刷版） | 1934-1768 |
| ISSN（电子版） | 2161-2927 |
会议
| 会议 | 40th Chinese Control Conference, CCC 2021 |
|---|---|
| 国家/地区 | 中国 |
| 市 | Shanghai |
| 时期 | 26/07/21 → 28/07/21 |
指纹
探究 'Action Recognition Algorithm based on 2D Human Pose Estimation Method' 的科研主题。它们共同构成独一无二的指纹。引用此
Yu, C., Chen, W., Li, Y., & Chen, C. (2021). Action Recognition Algorithm based on 2D Human Pose Estimation Method. 在 C. Peng, & J. Sun (编辑), Proceedings of the 40th Chinese Control Conference, CCC 2021 (页码 7366-7370). (Chinese Control Conference, CCC; 卷 2021-July). IEEE Computer Society. https://doi.org/10.23919/CCC52363.2021.9550204