UNIVERSITY OF KOBLENZ
Universitätsstraße 1
56070 Koblenz
With the increased tendency to use electronic media, the number of people who increasingly report neck pain is rising. Constantly looking at the smartphone, tablet or keyboard unintentionally forces the user to adopt a non-physiological upper body and head posture.
Biophotogrammetric methods can be used to make valid statements about posture weaknesses. Usually, such measurements are performed using markers attached to the subject's body, which provide conclusions about the current posture.
Our study presents a novel method to automate the detection of the prominent landmarks, such as the spinous process of the cervical vertebra C7 and the ear tragus in an RGB image. These points are used in the determination of the craniocervical angle.
In application, various image processing methods are used for this purpose together with a neural network Mediapipe. The method was evaluated on a dataset and the results of the method as well as a live demo will be presented at the CV day.
Visit us at the Computational Visualistics Day at the University of Koblenz and learn more about our research project "Smombie - Feedback about your personal neck posture".