Improved EfficientDet Algorithm for Basketball Players’ Upper Limb Movement Trajectory Recognition

Figures & data

Figure 1. Network structure diagram.

Figure 2. Structure of MBConv Block.

Figure 3. Spatial attention module diagram.

Figure 4. The improved BiFPN map.

Figure 5. Top-down feature fusion.

Table 1. List of compared algorithms.

Number	Provenance	Algorithm	Description
1	Proposed	Improved EfficientDet	The spatial attention mechanism is added to the backbone network of EfficientDet-D0, which can locate the limb action trajectory features in the images more accurately. Secondly, in the feature fusion network, in order to describe the high-frequency detail information lost by downsampling, the idea of Laplace pyramid is used to fuse the detail feature maps in the top-down fusion path, and cross-level connections are added to make full use of the feature information of different resolutions, so that the acquired high-level feature map information is richer. Finally, the whole network is trained using migration learning techniques and Adam optimizer.
2	Literature (Cavallari, Golodetz, and Lord 2019)	DeepPose	The pose estimation is formulated as a problem of direct regression of nodes, while the use of a cascade approach to obtain more accurate coordinate positions is proposed.
3	Literature (Silva et al. 2019)	CPM	CPM consists of a series of convolutional networks that repeatedly generate 2D confidence maps for each partial position. The experimental CPM network used here is of order 3.
4	Literature (Wang et al. 2019)	Stacked hourglass	Multiple Hourglass modules are integrated to capture the features of the image at different scale levels.

Table 2. PCP values of the components of the four methods on the LSP dataset %.

Method	Head	Trunk	Upper Arm	Forearm	Lap	Calf	Average
Proposed	86.2	93.0	62.7	47.3	80.4	72.7	70.8
			62.4	47.8	80.8	74.5
Literature (Cavallari, Golodetz, and Lord 2019)	81.4	91.0	56.5	38.3	77.7	71.4	66.1
			56.7	38.8	77.5	71.6
Literature (Silva et al. 2019)	92.3	95.2	85.2	77.6	86.3	82.9	85.3
			85.7	78.8	86.2	82.4
Literature (Wang et al. 2019)	89.9	95.8	83.4	75.8	86.0	79.6	83.5
			83.2	76.0	84.6	80.4

Table 3. Number of parameters and time to test 1000 times for the four methods.

Method	Number of participants	Time consumption/s
Proposed	2863370	7.41
Literature[24]	21695910	232.53
Literature[25]	16585375	178.30
Literature[26]	9716490	27.62

Figure 6. PCP values of the four methods on the upper limb posture dataset.

Table 4. Number of participants and time to test 1000 times for the four methods.

Method	Number of participants	Time consumption/s
Proposed	1636470	5.34
Literature[24]	21646720	23.80
Literature[25]	16572276	177.87
Literature[26]	9708790	26.92

Cavallari, T., S. Golodetz, and N. A. Lord. 2019. Real-Time RGB-D Camera pose estimation in novel scenes using a relocalisation cascade[J. IEEE Transactions on Pattern Analysis and Machine Intelligence 42 (10):2465–77. doi:10.1109/TPAMI.2019.2915068.

 PubMed Web of Science ®Google Scholar

Severin, C., A. Tackett, S. A, and C. L. Barnes. 2022. Three-Dimensional Kinematics in Healthy Older Adult Males during Golf Swings[J. Sports Biomechanics 21 (2):165–78. doi:10.1080/14763141.2019.1649452.

 PubMedGoogle Scholar

Chen, C., T. Wang, D. Li, and J. Hong. 2020. Repetitive assembly action recognition based on object detection and pose estimation[J. Journal of Manufacturing Systems 55:325–33. doi:10.1016/j.jmsy.2020.04.018.

 Web of Science ®Google Scholar

Silva, L. J. S., D. L. S. Silva, A. B. Raposo, L. Velho, and H. Côrtes Vieira Lopes. 2019. Tensorpose: Real-time pose estimation for interactive applications[J. Computers & Graphics 85:1–14. doi:10.1016/j.cag.2019.08.013.

 Web of Science ®Google Scholar

Wang, R., Z. Cao, X. Wang, Z. Liu, and X. Zhu. 2019. Human pose estimation with deeply learned multi-scale compositional models[J. IEEE Access 7:71158–66. doi:10.1109/ACCESS.2019.2919154.

 Web of Science ®Google Scholar