Intelligent Posture Monitoring and Correction System Using MediaPipe and Computer Vision

This presentation introduces an intelligent posture monitoring and correction system based on computer vision techniques. The system uses MediaPipe for real-time human pose estimation and OpenCV for image processing and body angle analysis. It continuously evaluates neck, shoulder, and back alignment to detect incorrect posture during prolonged sitting. The proposed solution addresses limitations of traditional methods by providing real-time feedback without requiring wearable devices. A progressive screen blurring mechanism is implemented to alert users and encourage immediate posture correction. The system operates locally on the user’s device, ensuring full data privacy and low computational cost. It achieves real-time performance exceeding 30 frames per second under normal conditions. Experimental results and related studies indicate high accuracy in posture detection, reaching up to 95% in similar approaches. The system is expected to reduce physical strain and decrease back and neck pain by approximately 35–45%. Overall, the solution contributes to improving productivity and preventing long-term musculoskeletal disorders.

This work was conducted at Arab International University (AIU), Syria.
The official website of the university is: https://www.aiu.edu.sy