Virtualização de avatares humanos com sensor Kinect para aplicações em realidade virtual

Abstract

The advent of virtual meetings, studies, work, conferences, debates, entertainment, and even physical activities in virtual environments is expected to grow in the future. The use of virtualized avatars allows for humanizing individuals in the digital world, making the present work a case study for reality virtualization and augmented reality. This work analyzes the utilization of a Kinect sensor for the construction and animation of a 3D virtual avatar with the ability to be incorporated into a virtual environment. The Kinect sensor is used to scan a human face and digitize it into a three-dimensional model. The 3D model is subsequently animated through facial capture by the Kinect sensor and interpretation of facial points of interest. Success has been achieved in digitally modeling the human face; however, the mesh created by the Kinect sensor lacks sufficient resolution for the construction of virtual bones used in animation. Therefore, a second high-definition model was used for animation.