Monte Carlo Hidden Markov Models Stanford University蒙特卡洛隐藏的马尔可夫模型斯坦福大学.pptVIP

Structure From Motion Sebastian Thrun, Gary Bradski, Daniel Russakoff Stanford CS223B Computer Vision /cs223b Structure From Motion (1) Structure From Motion (2) Structure From Motion (3) Structure From Motion Problem 1: Given n points pij =(xij, yij) in m images Reconstruct structure: 3-D locations Pj =(xj, yj, zj) Reconstruct camera positions (extrinsics) Mi=(Aj, bj) Problem 2: Establish correspondence: c(pij) Orthographic Camera Model Limit of Pinhole Model: The Affine SFM Problem Count # Constraints vs #Unknowns m camera poses n points 2mn point constraints 8m+3n unknowns Suggests: need 2mn ? 8m + 3n But: Can we really recover all parameters??? Points for Solving Affine SFM Problem m camera poses n points Need to have: 2mn ? 8m + 3n-12 Affine SFM The Rank Theorem Tomasi/Kanade 1992 Tomasi/Kanade 1992 Back To Projective Geometry Structure From Motion Problem 1: Given n points pij =(xij, yij) in m images Reconstruct structure: 3-D locations Pj =(xj, yj, zj) Reconstruct camera positions (extrinsics) Mi=(Aj, bj) Problem 2: Establish correspondence: c(pij) The Correspondence Problem Correspondence: Solution 1 Track features (e.g., optical flow) …but fails when images taken from widely different poses Correspondence: Solution 2 Start with random solution A, b, P Compute soft correspondence: p(c|A,b,P) Plug soft correspondence into SFM Reiterate See Dellaert/Seitz/Thorpe/Thrun 2003 Example Results: Cube Reconstruction with EM 3-D Structure Sebastian Thrun Stanford University CS223B Computer Vision * [Tomasi Kanade 92] [Tomasi Kanade 92] [Tomasi Kanade 92] Extrinsic Parameters Rotation Orthographic Projection Orthographic Projection Limit of Pinhole Model: Orthographic Projection How Many Parameters Can’t We Recover? nm m n 12 10 9 8 6 3 0 Place Your Bet! We can recover all but… The Answer is (at least): 12 Fix coordinate system by making p0=origin Proof: Rank Theorem: Q has rank 3 n elements 2m elements S