Year: 2013
Type: conference-paper
Abstract: Image-based modeling is becoming increasingly popular as a means to create realistic 3D digital models of realworld\nobjects. Applications range from games and e-commerce to virtual worlds and 3D printing. Most research\nin computer vision has concentrated on the precise reconstruction of geometry. However, in order to improve\nrealism and enable use in professional production pipelines digital models need a high-resolution texture map. In\nthis paper we present a novel system for creating detailed texture maps from a set of input images and estimated\n3D geometry. The solution uses a mesh segmentation and charting approach in order to create a low-distortion\nmesh parameterization suitable for objects of arbitrary genus. Texture maps for each mesh segment are created\nby back-projecting the best-fitting input images onto each surface segment, and smoothly fusing them together\nusing graph-cut techniques. We investigate the effect of different input parameters, and present results obtained for\nreconstructing a variety of different 3D objects from input images acquired using an unconstrained and uncalibrated\ncamera.
Authors: Hoang Minh Nguyen, Burkhard C. Wünsche, Patrice Delmas, Christof Lutteroth, Wannes van der Mark, Eugene Zhang
Institutions: University of Auckland, Oregon State University
Language: English
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2397006379