CS 532 3D Computer Vision

Computer vision addresses the image understanding problem; in other words, it aims to infer what was depicted in still images or video based on pixel intensity or color values. Never is the relationship between the depicted scene and images more explicit than in 3D computer vision that aims to extract 3D information from image and video data, as well as other modalities. This course will introduce students to concepts relating 2D images and 3D scenes including single and multiple- view geometry, structure from motion and 3D reconstruction. It will also cover processing of 3D data regardless of its origin starting from point sets and progressing to lines, polygons, Delaunay triangulations and Voronoi diagrams. Students will acquire in depth knowledge of 3D computer vision topics that have moved to the forefront for a broad range of applications in geospatial information systems (Google and Bing maps), robotics and driver assistance, 3D user interfaces (Microsoft Kinect), augmented reality and visual aids for people with impaired sight.

Credits

3

Prerequisite

Graduate Student or At Least Junior

Distribution

Computer Science Program

Typically Offered Periods

Fall Semester