Ethical Decision‐Making in Yoga Posture Detection through AI
Ishita Jain, Riya Srivastava, Vanshita Srivastava, Vanshika Sinha, Abhinav Juneja · 2025
Yoga is an age-old established practice that encompasses a wide range of postures called asanas with various health and wellness benefits to the person who practices them. Yoga helps with a person's physical and mental wellness. To achieve these benefits, one need to practice yoga postures regularly. But sometimes bad postures can lead to multiple injuries and accidents. To overcome such difficulties, a wide range of machine learning algorithms are integrated to solve such issues. Over the past years, there has been an increase in the usage of this technology for automating the whole process from yoga posture detection to correcting these poses so as to enhance the user's experience. This chapter presents a comprehensive study of applications of machine learning related to yoga posture detection using OpenCV and MediaPipe. This chapter explores various machine learning algorithms and methodologies used for yoga (asanas) posture detection. It uses various machine learning technologies like deep learning, supervised learning, pose detection and estimation models to achieve accurate posture. Furthermore, it examines the importance of selecting the required data, implementing different data augmentation methodologies and then optimizing these models to improve the accuracy of the posture detection. The chapter examines the ethical decision-making framework necessary to inculcate data privacy measures and maintain transparency of the usage of the application end users. The following chapter also discusses the challenges faced with respect to ethical issues, data privacy concerns, etc., while using artificial intelligence frameworks to identify real-time yoga poses. This comprehensive study provides insights into the current state of yoga posture detection using ML algorithms, emphasizing the technical aspects, practical implications and ethical considerations. We identify avenues for future research, such as development of user-friendly applications, integration of multimodal data sources, and exploration of adaptive feedback mechanisms. The findings of this study contribute to the growing knowledge aimed at enhancing the yoga practice through technology-enabled posture detection and analysis.