Research Article
Sports Pose Estimation Based on LSTM and Attention Mechanism
3 downloads
@INPROCEEDINGS{10.1007/978-3-030-63941-9_45, author={Chuanlei Zhang and Lixin Liu and Qihuai Xiang and Jianrong Li and Xuefei Ren}, title={Sports Pose Estimation Based on LSTM and Attention Mechanism}, proceedings={6GN for Future Wireless Networks. Third EAI International Conference, 6GN 2020, Tianjin, China, August 15-16, 2020, Proceedings}, proceedings_a={6GN}, year={2021}, month={1}, keywords={Sports pose estimation Two-branch multi-stage CNN LSTM-attention mechanism}, doi={10.1007/978-3-030-63941-9_45} }- Chuanlei Zhang
Lixin Liu
Qihuai Xiang
Jianrong Li
Xuefei Ren
Year: 2021
Sports Pose Estimation Based on LSTM and Attention Mechanism
6GN
Springer
DOI: 10.1007/978-3-030-63941-9_45
Abstract
In our life, we often need to estimate the accuracy of sports pose, which usually costs a lot of time and human resources. To solve the problem, we propose a LSTM-Attention model. In spatial dimension, we use the two-branch multi-stage CNN to extract human joints as features, which not only guarantees the real-time performance, but also ensures the accuracy. For the time dimension, the extracted joint features sequence is input into the LSTM-Attention model for training. In order to verify the effectiveness of our proposed method, we collected data for processing and trained with the proposed model. The experimental results show that our method has a high performance.
Copyright © 2020–2025 ICST
