creator Loghman, Maziar creator Mesmakhosroshahi, Maral creator Kim, Joohee A computer-implemented depth estimation method based on non-parametric Census transform with adaptive window patterns and semi-global optimization. A modified cross-based cost aggregation technique adaptively creates the shape of the cross for each pixel distinctly. In addition, a depth refinement algorithm fills holes within the estimated depth map using the surrounding background depth pixels and sharpens the object boundaries by exerting a trilateral filter to the generated depth map. The trilateral filter uses the curvature of pixels as well as texture and depth information to sharpen the edges. Submitted by Aric Ahrens (ahrens@iit.edu) on 2018-01-04T22:14:01Z No. of bitstreams: 2 license_rdf: 1089 bytes, checksum: 0a703d871bf062c5fdc7850b1496693b (MD5) 9754377.pdf: 308928 bytes, checksum: 4df3fff5aa55327d280dc068345150ec (MD5) Made available in DSpace on 2018-01-04T22:14:02Z (GMT). No. of bitstreams: 2 license_rdf: 1089 bytes, checksum: 0a703d871bf062c5fdc7850b1496693b (MD5) 9754377.pdf: 308928 bytes, checksum: 4df3fff5aa55327d280dc068345150ec (MD5) Previous issue date: 2017-09-05 Illinois Institute of Technology 2018-01-04 2017-09-05 9,754,377 http://hdl.handle.net/10560/4195 en CC0 1.0 Universal http://creativecommons.org/publicdomain/zero/1.0/ United States Patent Patent reformatted digital application/pdf No Copyright - United States http://rightsstatements.org/page/NoC-US/1.0/ Open Access