Document Type
Article
Abstract
In today’s world, the increased time people spend in front of their computers has been one of the main causes for neck and back pains. Especially, since the pandemic, it has been quite evident that slouching at home for long hours on hand-held devices and computers has led many people towards spinal pains and injuries. Backed with scientific research, it has been proven that these pains can be prevented with proper monitoring of the seated posture and taking breaks in between. This paper focuses on building a light-weight end-to-end system that monitors the user’s posture and provides feedback whenever it is necessary for them to fix their posture or take a break. Our system utilizes the most common devices: a webcam or a smartphone camera to capture input frames and a machine learning model to differentiate between good and bad postures with 98% accuracy while the user is seated. The newly developed pipeline helps the users in improving their posture without any additional cost or hardware.
Publication Date
7-11-2022
Language
English
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Recommended Citation
Kapoor, Rithik; Jaiswal, Ashish; and Makedon, Fillia, "Light-Weight Seated Posture Guidance System with Machine Learning and Computer Vision" (2022). Association of Computing Machinery Open Access Agreement Publications. 31.
https://mavmatrix.uta.edu/utalibraries_acmoapubs/31