Proceedings of the 1st Engineering Data Analytics and Management Conference (EAMCON 2025)

<Previous Article In Volume
Next Article In Volume>

Yoga Pose Detection and Correction Using MediaPipe Landmarks and a Deep Multi-Layer Perceptron Classifier

Authors
D. Santhadevi1, *, Srinivas Yerigera1, S. Aaditya1, Boddu Ramana Kumar Reddy1
1SCOPE, VIT-AP University, Amaravati, Andhra Pradesh, India
*Corresponding author. Email: santhadevi.d@vitap.ac.in
Corresponding Author
D. Santhadevi
Available Online 31 December 2025.
DOI
10.2991/978-94-6463-978-0_30How to use a DOI?
Abstract

Yoga is widely used for improving flexibility, strength, balance, and mental well-being; however, incorrect posture execution can reduce efficacy and increase injury risk. A lightweight, interpretable pose recognition and feedback system is presented that combines MediaPipe Pose landmark extraction with a deep Multi-Layer Perceptron (MLP) classifier. A curated dataset covering 85 asanas was created and expanded via targeted augmentation to 10,000 examples. The system includes a geometric correction layer that converts landmark configurations into precise alignment suggestions for commodity hardware. Detailed methodology, ablation studies, and deployment guidance are provided to support reproducibility and on-device use. The final model achieves 77.62% accuracy on a stratified test set, with robust performance on distinct poses and remaining confusions among visually similar variants.

Copyright
© 2025 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

<Previous Article In Volume
Next Article In Volume>
Volume Title
Proceedings of the 1st Engineering Data Analytics and Management Conference (EAMCON 2025)
Series
Advances in Engineering Research
Publication Date
31 December 2025
ISBN
978-94-6463-978-0
ISSN
2352-5401
DOI
10.2991/978-94-6463-978-0_30How to use a DOI?
Copyright
© 2025 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Cite this article

risenwbib
TY  - CONF
AU  - D. Santhadevi
AU  - Srinivas Yerigera
AU  - S. Aaditya
AU  - Boddu Ramana Kumar Reddy
PY  - 2025
DA  - 2025/12/31
TI  - Yoga Pose Detection and Correction Using MediaPipe Landmarks and a Deep Multi-Layer Perceptron Classifier
BT  - Proceedings of the 1st Engineering Data Analytics and Management Conference (EAMCON 2025)
PB  - Atlantis Press
SP  - 332
EP  - 351
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6463-978-0_30
DO  - 10.2991/978-94-6463-978-0_30
ID  - Santhadevi2025
ER  -