A virtual yoga instructor uses PoseNet which is a machine learning model that allows for Real-time Human Pose Estimation and overlay a skeleton stick figure on top. Pose estimation refers to computer vision techniques that detect human figures in images and videos so that one could determine, for example, where someone’s elbow shows up in an image. To be clear, this technology is not recognizing who is in an image — there is no personally identifiable information associated to pose detection. The algorithm is simply estimating where key body joints are. PoseNet main objective is to inspire more developers and makers to experiment and apply pose detection to their own unique projects.
The body is the most complex biochemical factory we know of & learning how to engineer it allows us to enter sense modalities at will. We are all caught up in the rat race. Our mind, soul, and body need to experience a deep fulfilling joy free from all the worldly tensions. The first solution that pops up is YOGA. yoga is associated with the culture and heritage of India. In Sanskrit, yoga means to unite and describes a way to live a healthy life. This may be a difficult practice not necessarily fun but the benefits from it are worth it. Our project is to motivate society to practice yoga by conceiving, designing, and building the prototype for a yoga instructor.
Human pose estimation is an important problem in the field of Computer Vision. Imagine being able to track a person’s every small movement and do a biomechanical analysis in real-time. The technology will have huge implications. Applications may including video surveillance, assisted living, advanced driver assistance systems (ADAS) and sports analysis.
Yoga has an ancient tradition of physical and mental training to improve well-being. Modern Hatha Yoga, which emphasizes physical conditioning and mental strength through training physical postures, has been growing in popularity over the last couple decades. However, the time and cost of getting to a yoga studio can be prohibitive. Still others would prefer to practice outside of the group setting. Here, we explore the development of YogAI. With pose recognition, we implement a smart assistant to provide corrective advice to guide practitioners. The goal is to help a user reduce stress, anxiety, and depression through a series of guide meditative poses.
Visit the Website for more info 😄