Description of Invention
Prolonged poor posture is a growing concern, leading to adverse health effects such as back pain, reduced concentration, and musculoskeletal issues. Traditional methods for posture correction often rely on physiotherapists and specialized equipment, which can be inconvenient and inaccessible. To address this, the Stand Posture Corrector project leverages PoseNet, a pre-trained deep learning model, to detect standing posture in real time using a device’s camera. The application employs TensorFlow and PoseNet technologies to identify key body points and provide immediate feedback to users when poor posture is detected. This real-time correction system is designed to enhance user health and well-being while maintaining a user-friendly interface that operates on standard devices with minimal hardware requirements. This study highlights the integration of advanced machine learning techniques into everyday applications, focusing on accuracy, real-time performance, and accessibility. By addressing the limitations of existing posture correction methods, this project aims to promote better posture habits, reduce health risks, and improve the quality of life for users across medical, fitness, and ergonomic fields.