Limits...
Iterative most-likely point registration (IMLP): a robust algorithm for computing optimal shape alignment.

Billings SD, Boctor EM, Taylor RH - PLoS ONE (2015)

Bottom Line: Our GTLS approach has improved accuracy, efficiency, and stability compared to prior methods presented for this problem and offers a straightforward implementation using standard least squares.We evaluate the performance of IMLP relative to a large number of prior algorithms including ICP, a robust variant on ICP, Generalized ICP (GICP), and Coherent Point Drift (CPD), as well as drawing close comparison with the prior anisotropic registration methods of GTLS-ICP and A-ICP.The performance of IMLP is shown to be superior with respect to these algorithms over a wide range of noise conditions, outliers, and misalignments using both mesh and point-cloud representations of various shapes.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer Science, Johns Hopkins University, Baltimore, MD, United States of America.

ABSTRACT
We present a probabilistic registration algorithm that robustly solves the problem of rigid-body alignment between two shapes with high accuracy, by aptly modeling measurement noise in each shape, whether isotropic or anisotropic. For point-cloud shapes, the probabilistic framework additionally enables modeling locally-linear surface regions in the vicinity of each point to further improve registration accuracy. The proposed Iterative Most-Likely Point (IMLP) algorithm is formed as a variant of the popular Iterative Closest Point (ICP) algorithm, which iterates between point-correspondence and point-registration steps. IMLP's probabilistic framework is used to incorporate a generalized noise model into both the correspondence and the registration phases of the algorithm, hence its name as a most-likely point method rather than a closest-point method. To efficiently compute the most-likely correspondences, we devise a novel search strategy based on a principal direction (PD)-tree search. We also propose a new approach to solve the generalized total-least-squares (GTLS) sub-problem of the registration phase, wherein the point correspondences are registered under a generalized noise model. Our GTLS approach has improved accuracy, efficiency, and stability compared to prior methods presented for this problem and offers a straightforward implementation using standard least squares. We evaluate the performance of IMLP relative to a large number of prior algorithms including ICP, a robust variant on ICP, Generalized ICP (GICP), and Coherent Point Drift (CPD), as well as drawing close comparison with the prior anisotropic registration methods of GTLS-ICP and A-ICP. The performance of IMLP is shown to be superior with respect to these algorithms over a wide range of noise conditions, outliers, and misalignments using both mesh and point-cloud representations of various shapes.

No MeSH data available.


Related in: MedlinePlus

Registration errors for registering a source shape containing outliers to a point-cloud target under large misalignment. (Experiment 5B).Source shapes were randomly generated from a mesh model of a human hip (Fig. 1A), misaligned by [30, 60] mm / degrees, and registered back to a point-cloud representation of the mesh. The test cases represent different noise models used to generate noise on the source shape (Table 4). Outliers were added to the source shape constituting (A): 5%, (B): 10%, (C): 20%, and (D): 30% of the source points. For each test case, 300 randomized trials were conducted, with successful registrations being used to compute an average target registration error (TRE). The error bars provide approximate standard deviations of the reported average TRE values. The proposed IMLP algorithm was evaluated relative to standard ICP [1], GICP [11], a robust variant of ICP [4], and CPD [20].
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4352012&req=5

pone.0117688.g007: Registration errors for registering a source shape containing outliers to a point-cloud target under large misalignment. (Experiment 5B).Source shapes were randomly generated from a mesh model of a human hip (Fig. 1A), misaligned by [30, 60] mm / degrees, and registered back to a point-cloud representation of the mesh. The test cases represent different noise models used to generate noise on the source shape (Table 4). Outliers were added to the source shape constituting (A): 5%, (B): 10%, (C): 20%, and (D): 30% of the source points. For each test case, 300 randomized trials were conducted, with successful registrations being used to compute an average target registration error (TRE). The error bars provide approximate standard deviations of the reported average TRE values. The proposed IMLP algorithm was evaluated relative to standard ICP [1], GICP [11], a robust variant of ICP [4], and CPD [20].

Mentions: The TRE achieved by each algorithm in this study is presented in Fig. 6 for the range of small misalignments (Experiment 5A) and in Fig. 7 for the range of large misalignments (Experiment 5B). Figs. 6 and 7 are each divided into four sub-figures (A-D) corresponding to sub-experiments (i-iv) of Experiments 5A and 5B, respectively, for each level of outliers. As in the prior studies, the TRE outcomes for each misalignment range are very similar. As seen in the figures, IMLP achieves large improvement in registration accuracy relative to the other algorithms for up to 20% outliers, even in comparison to CPD, which has a very effective outlier rejection capability. For the 30% outlier case, IMLP continues to provide accurate results and compares approximately equal to CPD. Compared to Robust ICP, IMLP is substantially more accurate in all test cases and frequently achieves less than half the registration error and below. As expected, standard ICP and GICP perform poorly, since they are non-robust techniques and do not include mechanisms to account for outliers. Robust ICP fairs much better than the non-robust methods for outlier compositions of 10% and below, but produces higher registration error than standard ICP for the outlier percentages above 10%.


Iterative most-likely point registration (IMLP): a robust algorithm for computing optimal shape alignment.

Billings SD, Boctor EM, Taylor RH - PLoS ONE (2015)

Registration errors for registering a source shape containing outliers to a point-cloud target under large misalignment. (Experiment 5B).Source shapes were randomly generated from a mesh model of a human hip (Fig. 1A), misaligned by [30, 60] mm / degrees, and registered back to a point-cloud representation of the mesh. The test cases represent different noise models used to generate noise on the source shape (Table 4). Outliers were added to the source shape constituting (A): 5%, (B): 10%, (C): 20%, and (D): 30% of the source points. For each test case, 300 randomized trials were conducted, with successful registrations being used to compute an average target registration error (TRE). The error bars provide approximate standard deviations of the reported average TRE values. The proposed IMLP algorithm was evaluated relative to standard ICP [1], GICP [11], a robust variant of ICP [4], and CPD [20].
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4352012&req=5

pone.0117688.g007: Registration errors for registering a source shape containing outliers to a point-cloud target under large misalignment. (Experiment 5B).Source shapes were randomly generated from a mesh model of a human hip (Fig. 1A), misaligned by [30, 60] mm / degrees, and registered back to a point-cloud representation of the mesh. The test cases represent different noise models used to generate noise on the source shape (Table 4). Outliers were added to the source shape constituting (A): 5%, (B): 10%, (C): 20%, and (D): 30% of the source points. For each test case, 300 randomized trials were conducted, with successful registrations being used to compute an average target registration error (TRE). The error bars provide approximate standard deviations of the reported average TRE values. The proposed IMLP algorithm was evaluated relative to standard ICP [1], GICP [11], a robust variant of ICP [4], and CPD [20].
Mentions: The TRE achieved by each algorithm in this study is presented in Fig. 6 for the range of small misalignments (Experiment 5A) and in Fig. 7 for the range of large misalignments (Experiment 5B). Figs. 6 and 7 are each divided into four sub-figures (A-D) corresponding to sub-experiments (i-iv) of Experiments 5A and 5B, respectively, for each level of outliers. As in the prior studies, the TRE outcomes for each misalignment range are very similar. As seen in the figures, IMLP achieves large improvement in registration accuracy relative to the other algorithms for up to 20% outliers, even in comparison to CPD, which has a very effective outlier rejection capability. For the 30% outlier case, IMLP continues to provide accurate results and compares approximately equal to CPD. Compared to Robust ICP, IMLP is substantially more accurate in all test cases and frequently achieves less than half the registration error and below. As expected, standard ICP and GICP perform poorly, since they are non-robust techniques and do not include mechanisms to account for outliers. Robust ICP fairs much better than the non-robust methods for outlier compositions of 10% and below, but produces higher registration error than standard ICP for the outlier percentages above 10%.

Bottom Line: Our GTLS approach has improved accuracy, efficiency, and stability compared to prior methods presented for this problem and offers a straightforward implementation using standard least squares.We evaluate the performance of IMLP relative to a large number of prior algorithms including ICP, a robust variant on ICP, Generalized ICP (GICP), and Coherent Point Drift (CPD), as well as drawing close comparison with the prior anisotropic registration methods of GTLS-ICP and A-ICP.The performance of IMLP is shown to be superior with respect to these algorithms over a wide range of noise conditions, outliers, and misalignments using both mesh and point-cloud representations of various shapes.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer Science, Johns Hopkins University, Baltimore, MD, United States of America.

ABSTRACT
We present a probabilistic registration algorithm that robustly solves the problem of rigid-body alignment between two shapes with high accuracy, by aptly modeling measurement noise in each shape, whether isotropic or anisotropic. For point-cloud shapes, the probabilistic framework additionally enables modeling locally-linear surface regions in the vicinity of each point to further improve registration accuracy. The proposed Iterative Most-Likely Point (IMLP) algorithm is formed as a variant of the popular Iterative Closest Point (ICP) algorithm, which iterates between point-correspondence and point-registration steps. IMLP's probabilistic framework is used to incorporate a generalized noise model into both the correspondence and the registration phases of the algorithm, hence its name as a most-likely point method rather than a closest-point method. To efficiently compute the most-likely correspondences, we devise a novel search strategy based on a principal direction (PD)-tree search. We also propose a new approach to solve the generalized total-least-squares (GTLS) sub-problem of the registration phase, wherein the point correspondences are registered under a generalized noise model. Our GTLS approach has improved accuracy, efficiency, and stability compared to prior methods presented for this problem and offers a straightforward implementation using standard least squares. We evaluate the performance of IMLP relative to a large number of prior algorithms including ICP, a robust variant on ICP, Generalized ICP (GICP), and Coherent Point Drift (CPD), as well as drawing close comparison with the prior anisotropic registration methods of GTLS-ICP and A-ICP. The performance of IMLP is shown to be superior with respect to these algorithms over a wide range of noise conditions, outliers, and misalignments using both mesh and point-cloud representations of various shapes.

No MeSH data available.


Related in: MedlinePlus