Action Recognition Method in Sports Video Shear Based on Fish Swarm Algorithm
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
This research offers a sports video action recognition approach based on the fish swarm algorithm in light ofthe low accuracy of existing sports video action recognition methods. A modified fish swarm algorithm isproposed to construct invariant features and decrease the dimension of features. Based on this algorithm, localfeatures and global features can be classified. The experimental findings on the typical sports action data setdemonstrate that the key details of sports action can be successfully retained by the dimensionality-reducedfusion invariant characteristics. According to this research, the average recognition time of the proposed methodfor walking, running, squatting, sitting, and bending is less than 326 seconds, and the average recognition rateis higher than 94%. This proves that this method can significantly improve the performance and efficiency ofonline sports video motion recognition.
Abstract
This research offers a sports video action recognition approach based on the fish swarm algorithm in light ofthe low accuracy of existing sports video action recognition methods. A modified fish swarm algorithm isproposed to construct invariant features and decrease the dimension of features. Based on this algorithm, localfeatures and global features can be classified. The experimental findings on the typical sports action data setdemonstrate that the key details of sports action can be successfully retained by the dimensionality-reducedfusion invariant characteristics. According to this research, the average recognition time of the proposed methodfor walking, running, squatting, sitting, and bending is less than 326 seconds, and the average recognition rateis higher than 94%. This proves that this method can significantly improve the performance and efficiency ofonline sports video motion recognition.
참고문헌 (10)
- Haoran Wei, "Simultaneous Utilization of Inertial and Video Sensing for Action Detection and Recognition in Continuous Action Streams", IEEE Sensors Journal, 20(11), pp.6055~6063 (2020), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Jian Xiong, "Object-Level Trajectories Based Fine-Grained Action Recognition in Visual IoT Applications", IEEE Access, 7, pp.103629~103638 (2019), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Omar Elharrouss, "A combined multiple action recognition and summarization for surveillance video sequences", Applied Intelligence, 51(2), pp.690~712 (2020), Springer Science and Business Media LLC[DOI]
- Fang Liu, "Exploring privileged information from simple actions for complex action recognition", Neurocomputing, 380, pp.236~245 (2020), Elsevier BV[DOI]
- Farhad Pourpanah, "A reinforced fuzzy ARTMAP model for data classification", International Journal of Machine Learning and Cybernetics, 10(7), pp.1643~1655 (2018), Springer Science and Business Media LLC[DOI]
- Petr Elias, "Understanding the limits of 2D skeletons for action recognition", Multimedia Systems, 27(3), pp.547~561 (2021), Springer Science and Business Media LLC[DOI]
- Yuri Yudhaswana Joefrie, "Multi-Label Multi-Class Action Recognition With Deep Spatio-Temporal Layers Based on Temporal Gaussian Mixtures", IEEE Access, 8, pp.173566~173575 (2020), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Jun Xie, "Attention adjacency matrix based graph convolutional networks for skeleton-based action recognition", Neurocomputing, 440, pp.230~239 (2021), Elsevier BV[DOI]
- Jun-Hwa Kim, "Action Recognition in Videos Using Pre-Trained 2D Convolutional Neural Networks", IEEE Access, 8, pp.60179~60188 (2020), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Dennis Ludl, "Enhancing Data-Driven Algorithms for Human Pose Estimation and Action Recognition Through Simulation", IEEE Transactions on Intelligent Transportation Systems, 21(9), pp.3990~3999 (2020), Institute of Electrical and Electronics Engineers (IEEE)[DOI]