Fitness Algorithm for Human Posture Detection and Correction
Garima Panwar, Deepshikha Bhargava, Harsh Khatter · 2024
This paper presents a comprehensive review of recent advancements in human posture detection and correction using machine learning techniques. Poor posture is a prevalent issue contributing to various musculoskeletal disorders and overall discomfort. Traditional methods of posture correction often rely on subjective assessments or costly specialized equipment. Machine learning algorithms have emerged as promising tools for addressing these challenges by automatically analyzing posture-related data and providing personalized feedback to users. A hybrid approach combining convolutional neural networks (CNNs) and recurrent neural networks (RNNs) is employed to analyze both static and dynamic aspects of posture. The CNNs extract spatial features from depth images, while the RNNs capture temporal dependencies from inertial measurement units (IMUs) data sequences. Once an improper posture is detected, the system provides real-time feedback to the user through visual, auditory, or haptic interfaces, guiding them toward correct posture alignment. Feedback mechanisms may include visual overlays, audio cues, or vibration patterns tailored to the user's preferences.