Computer vision, machine learning based monocular biomechanical and security analysis

Modern computer vision technologies have served to bridge the gap between contemporary scientific analysis and machine learning assisted digital processing. Within the field of biomechanics, applied strategies incorporating both conventional and machine assisted means have shown great success in augmenting the observations of electromyogram graphical sensors; albeit within the constraints of specialized, multiple-source arrays. The ongoing study represents an endeavor to utilize several distinct technologies to achieve similar results with the application of a single optical sensor. This paper documents the research, development, and implementation of a monocular feature extraction pipeline, designed for the intended use of supplementing modern athletic biomechanical analysis and the security of the data. We will examine the techniques presented by related works, and how we have implemented these strategies into our framework.