Efficient Camera Smoothing in Sequential Structure-from-Motion Using Approximate Cross-Validation.
Michela FarenzenaAdrien BartoliYoucef MezouarPublished in: ECCV (3) (2008)
Keyphrases
- structure from motion
- cross validation
- camera parameters
- model selection
- bundle adjustment
- feature tracking
- scene structure
- computer vision
- focal length
- missing data
- point cloud
- image sequences
- image based localization
- hyperparameters
- support vector
- camera motion
- training set
- relative pose
- regularization parameter
- camera positions
- point features
- camera pose
- scene depth
- multibody
- data sets
- line correspondences
- feature selection
- d scene
- camera calibration
- pinhole camera
- motion stereo