Smart Posture Guard using Machine Learning

In today’s digital age, much technology is applied that has caused a highly sedentary lifestyle. They are strongly linked with chronic back pain and musculoskeletal disorders. Poor posture and prolonged sitting cause great discomfort and fatigue that can cause long-term damage to the spine and surrounding muscles. Most people have poor postures because though there exist ergonomic solutions for this purpose, yet these mechanisms do not provide constant feedback in order to get rid of the improper alignment habit. This article discusses one approach to real-time posture detection by constantly analyzing the different postures and subsequently gives immediate feedbacks for the perfect alignment. This system uses machine learning and computer vision to capture the user's posture through a webcam, process it using the COCO dataset, identifies key body points, and uses a KNN model for classification as "Good" or "Bad." It notifies the user in real-time if it finds bad posture so that it can adjust on time. It also saves the posture data in user accounts so that it would be able to draw daily summaries, long term trends, and provide personalized feedback. The solution is designed to minimize health risks due to posture. This has helped people establish healthier sitting habits in an increasingly digital and screen-dependent world.