基於OpenPose特徵點評估儀隊動作序列

Quantitative Evaluation of Honor Guard’s Motion Sequence Using OpenPose Feature Points

吳宗諺(ZONG-YAN Wu)

指導教授 ： 張元翔

中原大學/電機資訊學院/資訊工程學系/碩士(2022年)

https://doi.org/10.6840/cycu202201360

摘要

近年疫情嚴峻的情勢下，許多學生儀隊無法在校園進行團體教學；而儀隊又依賴教學指導，在基礎動作訓練不準確的情況下容易造成運動傷害、特技動作難以訓練。所以更突顯訓練長以及教練的作用，在他們依循系統的指導下循序漸進的訓練才能使儀隊槍法做得更加精實、減少運動傷害的風險。
依靠人工智慧的發展，我們可以使用2D影像來偵測人體骨架特徵進而去比對動作的一致性、及各個關節點的位移對比；由此可知道自己與訓練長、教練動作的比對進而完善自身的動作。
本篇論文參考了美國卡內基梅隆大學(CMU)研究的Openpose 人體姿態AI模型作為人體關鍵特徵點的基礎，身為儀隊訓練長的動作為標準動作之基準。藉由Dynamic Time Warping來評斷動作在時間軸上不準確的地方，讓訓練者知道自身動作的不足處。

關鍵字

人工智慧 ； 人體骨架偵測 ； 動作比對 ； 時間動態規整法

並列摘要

Training of honor guards generally relies on guidance and practice. We present a video processing system for quantitative evaluation of honor guard’s motion sequence using OpenPose feature points. Our experimental results demonstrate the feasibility of automatic motion analysis between the trainer and the practitioner to yield useful feedback information for further improvement.

並列關鍵字

Motion Analysis ； OpenPose ； Pose Estimation ； Video Processing

參考文獻

[1] Z. Cao, G. Hidalgo, T. Simon, S. -E. Wei and Y. Sheikh, “OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 43, no. 1, pp. 172-186, 2021.

[2] S. Qiao, Y. Wang and J. Li, “Real-time human gesture grading based on OpenPose,” International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), pp. 1-6, 2017.

[3] G. Ning, P. Liu, X. Fan and C. Zhang, “A Top-Down Approach to Articulated Human Pose Estimation and Tracking,” ECCV Workshops, 2018.

[4] A. Viswakumar, V. Rajagopalan, T. Ray and C. Parimi, “Human Gait Analysis Using OpenPose,” Fifth International Conference on Image Information Processing (ICIIP), pp. 310-314, 2019.

[5] W. Lin and J. Ding, “Behavior detection method of OpenPose combined with Yolo network,” International Conference on Communications, Information System and Computer Engineering (CISCE), pp. 326-330, 2020.