Implementation of Computer Vision in Detecting Human Poses

Developing strong core muscles are important for children. Children with strong core muscles allow them to do any kinds of activities that mostly involve physical movement. The work tested on an education organisation that aims to strengthen core muscles by providing yoga-like poses. There are certified trainers that will coach the students along the way. However, mistakes could be made during the coaching because of different trainer's justification and whether the coaching processes were done correctly. Therefore, a solution is proposed to develop a computerized pose detector package which allows the trainers to improve the coaching with the students. The result is promising where the standardized pose could be implemented and compared to observed students' poses, however it is found that due to the uniqueness of the poses, it generates several unidentified results.

[1]  Yaser Sheikh,et al.  OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields , 2018, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Nobuo Funabiki,et al.  A Proposal of Yoga Pose Assessment Method Using Pose Detection for Self-Learning , 2019, 2019 International Conference on Advanced Information Technologies (ICAIT).

[3]  Richard Szeliski,et al.  Computer Vision - Algorithms and Applications , 2011, Texts in Computer Science.

[4]  Kehan Chen.  Sitting Posture Recognition Based on OpenPose , 2019 .

[5]  E. Arendt.  Core strengthening. , 2007, Instructional course lectures.

[6]  Wanli Ouyang,et al.  Towards Multi-Person Pose Tracking : Bottom-up and Top-down Methods , 2017 .

[8]  Ashwini Pansare,et al.  Match Pose - A System for Comparing Poses , 2019 .

[9]  Abhishek Gupta,et al.  Real-time Yoga recognition using deep learning , 2019, Neural Computing and Applications.

[10]  Varun Ramakrishna,et al.  Convolutional Pose Machines , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).