Abstract | ||
---|---|---|
Multi-view stereo (MVS) algorithms now produce recon- structions that rival laser range scanner accuracy. How- ever, stereo algorithms require textured surfaces, and there- fore work poorly for many architectural scenes (e.g., build- ing interiors with textureless, painted walls). This paper presents a novel MVS approach to overcome these limi- tations for Manhattan World scenes, i.e., scenes that con- sists of piece-wise planar surfaces with dominant direc- tions. Given a set of calibrated photographs, we first re- construct textured regions using an existing MVS algorithm, then extract dominant plane directions, generate plane hy- potheses, and recover per-view depth maps using Markov random fields. We have tested our algorithm on several datasets ranging from office interiors to outdoor buildings, and demonstrate results that outperform the current state of the art for such texture-poor scenes. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1109/CVPR.2009.5206867 | Miami, FL |
Keywords | Field | DocType |
Markov processes,image texture,laser ranging,stereo image processing,Manhattan-world stereo,Markov random field,architectural scene,calibrated photographs,depth maps,dominant plane directions,laser range scanner,multiview stereo algorithm,piecewise planar surfaces,plane hypothesis,textured regions,textured surfaces | Iterative reconstruction,Computer vision,Markov process,Computer graphics (images),Computer science,Markov random field,Stereopsis,Image texture,Markov chain,Ranging,Artificial intelligence,Scanner | Conference |
ISSN | ISBN | Citations |
1063-6919 | 978-1-4244-3992-8 | 94 |
PageRank | References | Authors |
3.15 | 17 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yasutaka Furukawa | 1 | 2388 | 88.91 |
Brian Curless | 2 | 8451 | 531.91 |
Steven M. Seitz | 3 | 8729 | 495.13 |
Richard Szeliski | 4 | 21300 | 2104.74 |