Human posture recognition is one of the most challenging tasks due to the variation in human appearance, changes in the background and illumination, additional noise in the frame, and diverse characteristics and amount of data generated. Aside from these, generating a high configuration for recognition of human body parts, occlusion, nearly identical parts of the body, variations of colors due to clothing, and other various factors make this task one of the hardest in computer vision. Therefore, these studies require high-computing devices and machines that could handle the computational load of this task. This study used a small-scale convolutional neural network and a smartphone built-in camera to recognize proper and improper sitting posture in a work-from-home setup. Aside from the recognition of body points, this study also utilized points’ distances and angles to help in recognition. Overall, the study was able to develop two objective datasets capturing the left and right side of the participants with the supervision and guidance of licensed physical therapists. The study shows accuracies of 85.18% and 92.07%, and kappas of 0.691 and 0.838, respectively. The system was developed, implemented, and tested in a work-from-home environment.

The sudden change in our workplace practices from face-to-face work to work from home setup due to the pandemic has brought positive and negative impacts on our overall health. In literature, the use of deep learning and specialized cameras in the estimation of the human pose is popular even if there is a need for high computational resources and complex models. For this purpose, this study developed an intelligent and interactive system utilizing a human estimation model with the use of distinct keypoint such as thoracic, thoraco lumbar, and lumbar points in the spine. An objective type of a dataset captured in a work from home environment with the knowledge and guidance of Licensed Physical Therapists to assess proper and improper sitting posture was developed. The study developed and implemented a small-scale convolutional network and low-cost smartphone camera to recognize body key points. Once all the feature points' locations were extracted, additional features such as cosine similarity and point distances were calculated. Next, feature selection and optimization were utilized to classify proper and improper sitting postures. As a result, the study developed (2) datasets and (2) models with an accuracy of 85.18 and 92.07% and kappa of 0.691 and 0.838 respectively.