Investigating the Construction of an Animated 3D Model Using Stereo Photogrammetry

3D modeling and motion capture are technologies that are widely used in many industries including virtual entertainment, clinical research, cinema and military. This technology is used to construct animated 3D models that can be controlled by human operators or designed to perform a series of tasks in a virtual environment. Other uses include scenario research and video gaming. Today, most 3D modeling techniques require a 3D designer to manually create the model. This has the potential to improve the quality and aesthetic of the model, and proves useful in designing video game characters, but is less accurate when the model is based on a real-world individual for the purpose of research. In most cases, 3D modeling and motion-capture are two separate operations. The animation is only mapped to the character in a later stage. This makes for longer working time, and a traditional motion-capture facility requires expensive equipment such as infra-red sensors and motion detection suits. Hence, in this paper, a method to create an animated 3D model in a single operation using photogrammetry technology is developed and evaluated. By combining motion-capture and 3D modeling in one operation, we can improve efficiency. Photogrammetric processing can recreate a 3D model with accurate measurements and details, which is advantageous in scientific research. Nonetheless, the quality of the model is influenced by the type of motions the character carried out as well as the setting and condition of the equipment.