Surface reconstruction and 3D modeling is a challenging task, which has been explored for decades by the computer vision, computer graphics, and machine learning communities. It is fundamental to many applications such as robot navigation, animation and scene understanding, industrial control and medical diagnosis. In this dissertation, I take advantage of the consumer depth sensors for surface reconstruction. Considering its limited performance on capturing detailed surface geometry, a depth enhancement approach is proposed in the first place to recovery small and rich geometric details with captured depth and color sequence. In addition to enhancing its spatial resolution, I present a hybrid camera to improve the temporal resolution of consumer depth sensor and propose an optimization framework to capture high speed motion and generate high speed depth streams. Given the partial scans from the depth sensor, we also develop a novel fusion approach to build up complete and watertight human models with a template guided registration method. Finally, the problem of surface reconstruction for non-Lambertian objects, on which the current depth sensor fails, is addressed by exploiting multi-view images captured with a hand-held color camera and we propose a visual hull based approach to recovery the 3D model.
| Identifer | oai:union.ndltd.org:uky.edu/oai:uknowledge.uky.edu:cs_etds-1097 |
| Date | 01 January 2019 |
| Creators | Zuo, Xinxin |
| Publisher | UKnowledge |
| Source Sets | University of Kentucky |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Theses and Dissertations--Computer Science |
Page generated in 0.0032 seconds