DOI QR코드

DOI QR Code

홈 트레이닝을 위한 운동 동작 분류 및 교정 시스템

Pose Classification and Correction System for At-home Workouts

  • Kang, Jae Min (School of Computer Science and Engineering, Pusan National University) ;
  • Park, Seongsu (Department of Information Convergence Engineering, Pusan National University) ;
  • Kim, Yun Soo (Department of Information Convergence Engineering, Pusan National University) ;
  • Gahm, Jin Kyu (School of Computer Science and Engineering, Pusan National University)
  • 투고 : 2021.07.06
  • 심사 : 2021.07.25
  • 발행 : 2021.09.30

초록

홈 트레이닝을 하는 사람들은 전문적인 대면 지도가 없기 때문에 잘못된 자세로 동작을 하여 신체에 무리가 올 수 있다. 본 연구에서는 자세 예측 모델과 다층 퍼셉트론을 이용하여 사용자의 자세를 교정 해주는 "영상 데이터 기반 동작 분류 및 자세 교정 시스템"을 제안한다. 자세 예측 모델로 뼈대 정보를 예측한 후 심층 신경망을 이용하여 어떤 운동 동작인지를 분류한 뒤, 올바른 관절의 각도를 알려주며 교정이 이루어진다. 이 과정에서 동작 분류 모델의 성능을 향상시키기 위해 연속적인 프레임들의 결과를 고려하는 투표 알고리즘을 적용하였다. 다층 퍼셉트론 기반 모델을 자세 분류 모델로 사용했을 때 0.9의 정확도를 가진다. 그리고 투표 알고리즘을 통해 분류 모델의 정확도는 0.93으로 향상된다.

There have been recently an increasing number of people working out at home. However, many of them do not have face-to-face guidance from experts, so they cannot effectively correct their wrong pose. This may lead to strain and injury to those doing home training. To tackle this problem, this paper proposes a video data-based pose classification and correction system for home training. The proposed system classifies poses using the multi-layer perceptron and pose estimation model, and corrects poses based on joint angels estimated. A voting algorithm that considers the results of successive frames is applied to improve the performance of the pose classification model. Multi-layer perceptron model for post classification shows the highest accuracy with 0.9. In addition, it is shown that the proposed voting algorithm improves the accuracy to 0.93.

키워드

과제정보

This work was supported by Institute of Information & communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (MSIT) (No. 2020-0-01450, Artificial Intelligence Convergence Research Center [Pusan National University]); and the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIT) (No. NRF-2020R1C1C1008362.

참고문헌

  1. J. Y. Kwon, I. H. Jo, and K. K. Choi, "Home Training Awareness Analysis Using Big Data: Focusing on the Spread of COVID-19," Journal of the Korean Physical Science Association, vol. 30, no. 2, pp. 447-459, 2020.
  2. J. H. Oh, J. W. Oh, and K. M. Jo, "A study on the continuation intention of one-person media YouTube home training by applying the late acceptance model," Journal of the Korean Convergence Society, vol. 10, no. 2, pp. 183-193, 2019.
  3. S. Chen and R. R. Yang, "Pose Trainer: correcting exercise posture using pose estimation," arXiv preprint arXiv. 2006.11718, 2020.
  4. D. J. Berndt and J. Clifford, "Using dynamic time warping to find patterns in time series," KDD workshop, vol. 10, no. 16, pp. 359-370, 1994.
  5. A. Talal, "Chen, Chen. Recognizing Exercises and Counting Repetitions in Real Time," arXiv preprint arXiv:2005.03194, 2020.
  6. J. H. Lee. (2019). Don't do pull-ups and push-ups [Internet]. Available: https://www.donga.com/news/It/article/all/20190620/96076041/1.
  7. B. Parhami, "Voting algorithms," IEEE transactions on reliability, vol. 43, no. 4, pp. 617-629, 1994. https://doi.org/10.1109/24.370218
  8. Carnegie Mellon University. openpose [Internet]. Available: https://github.com/CMU-Perceptual-Computing-Lab/openpose.
  9. Z. Cao, T. Simon, S. E. Wei, G. Hidalgo, 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, 2019.
  10. A. Mackiewicz and W. Ratajczak, "Principal components analysis (PCA)," Computers & Geosciences, vol. 19, no. 3, pp. 303-342, 1993. https://doi.org/10.1016/0098-3004(93)90090-R
  11. H. R Yoo and B. H Lee, "An OpenPose-based Child Abuse Decision System using Surveillance Video," Journal of the Korea Institute of Information and Communication Engineering, vol. 23, no. 3, pp.282-290, 2019. https://doi.org/10.6109/JKIICE.2019.23.3.282