Download presentation
1
Registration of Point Cloud Data from a Geometric Optimization Perspective
Niloy J. Mitra1, Natasha Gelfand1, Helmut Pottmann2, Leonidas J. Guibas1 1 Stanford University 2 Vienna University of Technology
2
Registration Problem Given Q P model data Assume Q is a part of P.
Two point cloud data sets P (model) and Q (data) sampled from surfaces P and Q respectively. Q P model data Assume Q is a part of P. Registration of PCD
3
Registration Problem Given Goal Q P data model
Two point cloud data sets P and Q. Goal Register Q against P by minimizing the squared distance between the underlying surfaces using only rigid transforms. Q P data model Registration of PCD
4
Registration Problem Given Goal Contributions
Two point cloud data sets P and Q. Goal Register Q against P by minimizing the squared distance between the underlying surfaces using only rigid transforms. Contributions use of second order accurate approximation to the squared distance field. no explicit closest point information needed. proposed algorithm has good convergence properties. Registration of PCD
5
Related Work Iterated Closest Point (ICP) Matching point clouds
point-point ICP [ Besl-McKay ] point-plane ICP [ Chen-Medioni ] Matching point clouds based on flow complex [ Dey et al. ] based on geodesic distance [ Sapiro and Memoli ] MLS surface for PCD [ Levin ] Registration of PCD
6
Notations Registration of PCD
7
Notations Registration of PCD
8
Squared Distance Function (F)
x Registration of PCD
9
Squared Distance Function (F)
x Registration of PCD
10
Registration Problem Rigid transform that takes points
Our goal is to solve for, An optimization problem in the squared distance field of P, the model PCD. Registration of PCD
11
Registration Problem Our goal is to solve for, Optimize for R and t.
Registration of PCD
12
Overview of Our Approach
Construct approximate such that, to second order. Linearize a. Solve to get a linear system. Apply a to data PCD (Q) and iterate. Registration of PCD
13
Registration in 2D Quadratic Approximant Registration of PCD
14
Registration in 2D Quadratic Approximant Linearize rigid transform
Registration of PCD
15
Registration in 2D Quadratic Approximant Linearize rigid transform
Residual error Registration of PCD
16
Depends on F+ and data PCD (Q).
Registration in 2D Minimize residual error Depends on F+ and data PCD (Q). Registration of PCD
17
Registration in 2D Minimize residual error Solve for R and t.
Apply a fraction of the computed motion F+ valid locally Step size determined by Armijo condition Fractional transforms [Alexa et al. 2002] Registration of PCD
18
Minimize to get a linear system
Registration in 3D Quadratic Approximant Linearize rigid transform Residual error Minimize to get a linear system Registration of PCD
19
Approximate Squared Distance
valid in the neighborhood of x Two methods for estimating F d2Tree based computation On-demand computation Registration of PCD
20
F(x, P) using d2Tree [ Leopoldseder et al. 2003]
A kd-tree like data structure for storing approximants of the squared distance function. Each cell (c) stores a quadratic approximant as a matrix Qc. Efficient to query. [ Leopoldseder et al. 2003] Registration of PCD
21
F(x, P) using d2Tree A kd-tree like data structure for storing approximants of the squared distance function. Each cell (c) stores a quadratic approximant as a matrix Qc. Efficient to query. Simple bottom-up construction Pre-computed for a given PCD. Closest point information implicitly embedded in the squared distance function. Registration of PCD
22
Example d2trees 2D 3D Registration of PCD
23
Approximate Squared Distance
Y For a curve Y, [ Pottmann and Hofer 2003 ] Registration of PCD
24
Approximate Squared Distance
For a curve Y, For a surface F, [ Pottmann and Hofer 2003 ] Registration of PCD
25
On-demand Computation
Given a PCD, at each point p we pre-compute, a local frame normal principal direction of curvatures radii of principal curvature (r1 and r2) Registration of PCD
26
On-demand Computation
Given a PCD, at each point p we pre-compute, a local frame normal principal direction of curvatures radii of principal curvature (r1 and r2) Estimated from a PCD using local analysis covariance analysis for local frame quadric fitting for principal curvatures Registration of PCD
27
On-demand Computation
Given a point x, nearest neighbor (p) computed using approximate nearest neighbor (ANN) data structure where j = d/(d-j) if d < 0 otherwise. Registration of PCD
28
Iterated Closest Point (ICP)
Find correspondence between P and Q. closest point (point-to-point). tangent plane of closest point (point-to-plane). Solve for the best rigid transform given the correspondence. Registration of PCD
29
ICP in Our Framework Point-to-point ICP (good for large d)
Point-to-plane ICP (good for small d) Registration of PCD
30
Convergence Properties
Gradient decent over the error landscape Gauss -Newton Iteration Zero residue problem (model and data PCD-s match) Quadratic Convergence For fractional steps, Armijo condition used Damped Gauss-Newton Iteration Linear convergence can be improved by quadratic motion approximation (not currently used) Registration of PCD
31
Convergence Funnel Set of all initial poses of the data PCD with respect to the model PCD that is successfully aligned using the algorithm. Desirable properties broad stable Registration of PCD
32
Convergence Funnel Translation in x-z plane. Rotation about y-axis.
Converges Does not converge Registration of PCD
33
Convergence Funnel Plane-to-plane ICP Our algorithm
Registration of PCD
34
Convergence Rate I Bad Initial Alignment Registration of PCD
35
Good Initial Alignment
Convergence Rate II Good Initial Alignment Registration of PCD
36
Partial Alignment Starting Position Registration of PCD
37
Partial Alignment After 6 iterations Registration of PCD
38
Partial Alignment Different sampling density After 6 iterations
Registration of PCD
39
Future Work Partial matching Global registration Non-rigid transforms
Registration of PCD
40
Questions? Registration of PCD
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.