Projecive structure from motion从运动射影结构.pptVIP

Projective structure from motion Marc Pollefeys COMP 256 Last class: Affine camera The affine projection equations are Orthographic factorization The orthographic projection equations are where Orthographic factorization Factorize m through singular value decomposition An affine reconstruction is obtained as follows Tentative class schedule Bundle adjustment - refining structure and motion Minimize reprojection error Maximum Likelyhood Estimation (if error zero-mean Gaussian noise) Huge problem but can be solved efficiently (exploit sparseness) Bundle adjustment Developed in photogrammetry in 50′s Non-linear least squares Linear approximation of residual allows quadratic approximation of sum- of-squares Bundle adjustment Jacobian of has sparse block structure cameras independent of other cameras, points independent of other points Bundle adjustment Eliminate dependence of camera/motion parameters on structure parameters Note in general 3n 11m Sequential SfM Initialize motion from two images Initialize structure For each additional view Determine pose of camera Refine and extend structure Refine structure and motion Initial projective camera motion Choose P and P′compatible with F Reconstruction up to projective ambiguity Initializing projective structure Reconstruct matches in projective frame by minimizing the reprojection error Projective pose estimation Infere 2D-3D matches from 2D-2D matches Compute pose from (RANSAC,6pts) Refining and extending structure Refining structure Extending structure 2-view triangulation Refining structure and motion use bundle adjustment Hierarchical structure and motion recovery Compute 2-view Compute 3-view Stitch 3-view reconstructions Merge and refine reconstruction Metric structure and motion use self-calibration (see next class) Dealing with dominant planes Farmhouse 3D models Application: video augmentation Next class: Camera calibration (and self-calibration) Non-iterati