Manuscript Title:

COMPUTER-ASSISTED TRAINING SYSTEM FOR YOGA AND PILATES USING COMPUTER VISION METHODS

Author:

STEVE BOBY GEORGE, Dr. JESWIN ROY DCOUTH, SANJAY JAIMY, EDWIN DAJI, AGNEL ANTONY

DOI Number:

DOI:10.17605/OSF.IO/FXSQE

Published : 2022-06-10

About the author(s)

1. STEVE BOBY GEORGE - Albertian Institute of Science and Technology Cochin University P. O., Kochi, Kerala, India.
2. Dr. JESWIN ROY DCOUTH - Albertian Institute of Science and Technology Cochin University P. O., Kochi, Kerala, India.
3. SANJAY JAIMY - Albertian Institute of Science and Technology Cochin University P. O., Kochi, Kerala, India.
4. EDWIN DAJI - Albertian Institute of Science and Technology Cochin University P. O., Kochi, Kerala, India.
5. AGNEL ANTONY - Albertian Institute of Science and Technology Cochin University P. O., Kochi, Kerala, India.

Full Text : PDF

Abstract

In recent times after the COVID-19 outbreak. People are subjected to being confined at their homes and they are finding it extremely difficult to practice yoga or pilates without a coach beside them. Yoga and pilates practitioners must pay high fees for a personal trainer who would essentially help them identify their mistakes and correct them in real time. The idea of a coach or trainer monitoring the practitioner would improve his/her's output tremendously. To minimize the need for a trainer, the proposed system would act as a trainer and identify the user's mistakes and correct them on the go. This would solve the problem of health and body ailments due to incorrect postures and would decrease the expense borne by the practitioners drastically.


Keywords

MoveNet, MediaPipe, Keypoints, Feature Vector, TensorFlow.js.