Download Project Document/Synopsis
Yoga, a centuries-old practice that is originally from India but is globally famous for its numerous spiritual, corporeal, and mental benefits is a type of exercise with complex postures. The problem with yoga is that, like any other exercise, it is critical to practice it correctly because any incorrect position during a yoga session can be ineffectual and potentially inconvenient. This necessitates the presence of a trainer to supervise the meeting and correct the individual’s stance. Since not every client approaches or has access to a trainer, a computerized reasoning-based application might be used to detect yoga poses and provide customized feedback to help people improve their structure.
Our Yoga Pose Detection System is designed and developed to recognize yoga stances and respond with a customized response to help users improve their postures. Our system will detect various yoga poses, namely Chair, Cobra, Dog, Shoulder Stand, Triangle, Tree, Warrior and No Pose.
Our Yoga Pose Detection System consist of 1 module: User. They can either upload a picture of a Yoga pose or pose directly in front of a camera and the system will automatically detect and show the name of the yoga pose.
The system will detect 8 types of Yoga poses Chair, Cobra, Dog, Shoulder Stand, Triangle, Tree, Warrior and No Pose. The libraries that are used in this project are OpenCV, dlib, OpenPose and MediaPipe. OpenPose is the first real-time multi-person system to jointly detect the human body, foot, hand, and facial key points on single images. The framework used in this project is Django. The Front End involves Html, CSS and JavaScript. The Back End involves MySQL Database.
Advantages
- The system is easy to maintain.
- It is user-friendly.
- The system successfully helps to identify Yoga poses.
- It aims to help users improve their poses.