International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 09 Issue: 04 | Apr 2022 www.irjet.net p-ISSN: 2395-0072 © 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 48 Virtual Yoga System Using Kinect Sensor Praful Wagh 1 , Sachin Patil 2 , Rushikesh Shrivastav 3 , Shubham Bachhav 4 [1,2,3,4] Students, Department of Computer Engineering, Loknete Gopinathji Munde Institute of Engineering Education & Research, Nashik, Maharashtra, India. ---------------------------------------------------------------------***--------------------------------------------------------------------- Abstract - Physical Yoga has become a big part of life for patients suffering from orthopedic injuries. But as time goes by, patients tend to get tired and demotivated due to repetitive and tedious exercises the recognition of poses is a field of investigation that takes incredible significance for oneself preparing in different sports. Due to its body tracking and depth sensor, Kinect provides a low-cost option for recognising Yoga positions. We propose an interactive system for perceiving a few postures for learning Yoga that will be characterised by a level of difficulty and coordinated with command voices to imagine the guidelines and pictures about the stances to be executed in this research. It also allows a therapist to customize exercises according to the specific needs and challenges of individual patients. The Microsoft Kinect sensor recognises skeleton and joint positions and provides three-dimensional information about the user's body. This enables an immersive and natural interaction between the user and the system. We use the Microsoft Kinect sensor to detect and track the movements of user and speech out message form system. Keywords: Kinect Sensor, Motion Capture, Virtual Physical Therapy, Serious Game, Rehabilitation, Interface Video Game, Interactive Technologies, Movement Recognition. 1. INTRODUCTION Physical therapy are optimal when assessment, monitoring, adherence to the therapy program and patient engagement can be achieved. Different procedures are involved in standard physical therapy and rehabilitation practise. These processes are usually intensive, time consuming, dependent on the expertise of the therapist and implies the collaboration of the patient who is usually asked to perform the therapy multiple times at home with no supervision. Injury or disease to the bone, articular, or soft tissue structures of the joints can result in discomfort, as well as a reduction in mobility, strength, stability, and functional use of the upper and lower extremities body traditional methods of rehabilitation is quite tedious and boring. As exercises are to be done for a long period of time, patients tends to loose interest over time because of monotonous exercises routines. Also, traditional physiotherapy involve use of goniometer which is a very old device used to measure degrees of flexion and extension of joints but it is inaccurate. The Microsoft Kinect Sensor has a number of advanced sensing components. It has a depth sensor, a colour camera, and a four-microphone array for full-body 3D motion capture, facial recognition, and voice recognition, among other features. A Kinect-based system can help patients do rehabilitation exercises correctly, promote patient accountability, and allow clinicians to rectify exercise faults. All these features allow to develop an integrated Human Computer Interaction (HCI) system in which happens as an open-ended dialog between the user and the computer. This system also provides some command voices so that the user can interact with the program such as the possibility of changing different Yoga poses instructions for a correct position. The System will prepare yoga program for the person. User should begin the system and kinect sensing element to perform the exercise. The voice instruction and, as a result, the action taken by the user will be the device's input. The kinect sensing element can locate the figure's twenty body points and relay the information to the system. The system is capable of calculating both RGB and depth images. The system may then calculate the movement and compare it to the rule-based instruction. If a user performs well, the system will reward them and direct them to the next step instruction to perform next yoga poses. Also, we can make weekly report of yoga exercise. Microsoft has announced new sensors that can establish watershed in RGB-D cameras for use with gaming consoles. There are three types of sensors: RGB, audio, and depth. Both of them perform important responsibilities, such as detecting movement, allowing users to play games using their bodies as a controller, and identifying player sounds. This Microsoft Kinect development also aided other computer vision applications like robotics and action identification. In Kinect sensor the arrangement of the infrared (IR) projector, the color camera, and the IR camera. The IR projector is paired with the IR camera, which is a monochrome complementary metal oxide semiconductor (CMOS) sensor, to form the depth sensor. The depth-sensing technology is licensed from the Israeli company Prime Sense. Although the particular technology isn't revealed, it is based on the notion of structured light. The IR projector is a series of IR dots created by an infrared laser passing through a diffraction grating. Gesture recognition system has been divided into five different phases. They are data acquisition, segmentation and pre-processing, feature extraction and finally the recognition. Data acquisition involves capturing