Abstract | ||
---|---|---|
Estimating the absolute pose of a camera relative to a 3D representation of a scene is a fundamental step in many geometric Computer Vision applications. When the camera is calibrated, the pose can be computed very efficiently. If the calibration is unknown, the problem becomes much harder, resulting in slower solvers or solvers requiring more samples and thus significantly longer run-times for RANSAC. In this paper, we challenge the notion that using minimal solvers is always optimal and propose to compute the pose for a camera with unknown focal length by randomly sampling a focal length value and using an efficient pose solver for the now calibrated camera. Our main contribution is a novel sampling scheme that enables us to guide the sampling process towards promising focal length values and avoids considering all possible values once a good pose is found. The resulting RANSAC variant is significantly faster than current state-of-the-art pose solvers, especially for low inlier ratios, while achieving a similar or better pose accuracy. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1007/978-3-319-10593-2_54 | COMPUTER VISION - ECCV 2014, PT IV |
Keywords | Field | DocType |
RANSAC, n-point-pose (PnP), camera pose estimation | Computer vision,Sampling process,RANSAC,Computer science,3D pose estimation,Focal length,Artificial intelligence,Sampling (statistics),Solver,Calibration,Sampling scheme | Conference |
Volume | ISSN | Citations |
8692 | 0302-9743 | 9 |
PageRank | References | Authors |
0.47 | 25 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Torsten Sattler | 1 | 704 | 34.68 |
Chris Sweeney | 2 | 101 | 7.42 |
Marc Pollefeys | 3 | 7671 | 475.90 |