posted on 2025-08-08, 12:16authored byMichael Hu Yang
Current methods of 3D shape acquisition do not take into account the size of the object unless expensive laser scanning equipment is used. This thesis provides details on the design and implementation of creating a 3D model to size for 3D printing. The thesis explores a pipeline using depth information gathered from a Microsoft Kinect and various open source software and freeware. The algorithms and techniques that are utilized to create the 3D models are also described. VisualSFM's point cloud generator, Meshlab's Poisson surface reconstructor, and Microsoft's 3D Builder are some of the pipeline components used in the research. Results obtained at different stages of the process will be presented and discussed.