Calculation of 3-D pose of a known object in a single perspective view

Dissertation

Kari Pehkonen

Research output: ThesisDissertationMonograph

Abstract

This thesis describes an algorithm for 3-D pose calculation of known objects from a single perspective view. The pose calculation algorithm consists of four main stages: pose candidate generation, candidate verification, optimization and reliability analysis. In the first stage, pose candidates are generated by matching model and image triangles. An important sub-problem, the so-called 3-point perspective pose estimation problem, is solved numerically. Pose candidates are verified by matching model and image features. The first accepted candidate is then optimized. Three optimization methods are considered: Newton-Raphson, simulated annealing and robust M-estimate using Lorenzian distribution. The reliability estimates are based on covariance analysis. Uncertainties due to both model and image noise are taken into account. A structure optimization method to reduce modeling noise is also presented. The algorithm was implemented on two parallel computers, a Connection Machine-2 (CM-2) and a Hathi-2/16. The Connection Machine is an SIMD machine whereas the Hathi-2/16 is based on the MIMD architecture. To evaluate the speed and accuracy of the algorithm, several experiments were performed using both synthetic and real images. In simulations, the accuracy of the calculated pose, the validity of uncertainty estimates and the effect of dynamic modeling were studied using a polyhedral test object whose maximum distance between any two vertices was 175 mm. Simulations showed that for the distance range from 0.5 m to 2.75 m, the standard deviation of the rotation error ranged from 0.5 degrees to 3.5 degrees when the standard deviation of image and model noise was 1.5 pixels and 1.5 mm, respectively. For the x-and y-directions the errors ranged from less than 2 mm up to 10 mm, and for the z-direction, from a few millimeters up to 75 mm. Much better results were achieved at the lower noise levels. The processing times on both the machines were approximately the same. On the Hathi-2/16, when all the 16 processors were used, the pose calculation time ranged from 0.376 s up to 0.814 s. For ease of programming the Hathi-2/16 was better than the CM-2, and it was also considerably cheaper.
Original languageEnglish
QualificationDoctor Degree
Awarding Institution
  • University of Oulu
Supervisors/Advisors
  • Pietikäinen, Matti, Supervisor, External person
Award date18 Dec 1992
Place of PublicationEspoo
Publisher
Print ISBNs951-38-4240-1
Publication statusPublished - 1992
MoE publication typeG4 Doctoral dissertation (monograph)

Fingerprint

Reliability analysis
Simulated annealing
Pixels
Processing
Experiments
Uncertainty

Keywords

  • 3d analysis
  • calculation
  • parallel computing
  • modelling

Cite this

Pehkonen, K. (1992). Calculation of 3-D pose of a known object in a single perspective view: Dissertation. Espoo: VTT Technical Research Centre of Finland.
Pehkonen, Kari. / Calculation of 3-D pose of a known object in a single perspective view : Dissertation. Espoo : VTT Technical Research Centre of Finland, 1992. 107 p.
@phdthesis{59ccfe94c3b44846be43c2001057be02,
title = "Calculation of 3-D pose of a known object in a single perspective view: Dissertation",
abstract = "This thesis describes an algorithm for 3-D pose calculation of known objects from a single perspective view. The pose calculation algorithm consists of four main stages: pose candidate generation, candidate verification, optimization and reliability analysis. In the first stage, pose candidates are generated by matching model and image triangles. An important sub-problem, the so-called 3-point perspective pose estimation problem, is solved numerically. Pose candidates are verified by matching model and image features. The first accepted candidate is then optimized. Three optimization methods are considered: Newton-Raphson, simulated annealing and robust M-estimate using Lorenzian distribution. The reliability estimates are based on covariance analysis. Uncertainties due to both model and image noise are taken into account. A structure optimization method to reduce modeling noise is also presented. The algorithm was implemented on two parallel computers, a Connection Machine-2 (CM-2) and a Hathi-2/16. The Connection Machine is an SIMD machine whereas the Hathi-2/16 is based on the MIMD architecture. To evaluate the speed and accuracy of the algorithm, several experiments were performed using both synthetic and real images. In simulations, the accuracy of the calculated pose, the validity of uncertainty estimates and the effect of dynamic modeling were studied using a polyhedral test object whose maximum distance between any two vertices was 175 mm. Simulations showed that for the distance range from 0.5 m to 2.75 m, the standard deviation of the rotation error ranged from 0.5 degrees to 3.5 degrees when the standard deviation of image and model noise was 1.5 pixels and 1.5 mm, respectively. For the x-and y-directions the errors ranged from less than 2 mm up to 10 mm, and for the z-direction, from a few millimeters up to 75 mm. Much better results were achieved at the lower noise levels. The processing times on both the machines were approximately the same. On the Hathi-2/16, when all the 16 processors were used, the pose calculation time ranged from 0.376 s up to 0.814 s. For ease of programming the Hathi-2/16 was better than the CM-2, and it was also considerably cheaper.",
keywords = "3d analysis, calculation, parallel computing, modelling",
author = "Kari Pehkonen",
note = "Project code: TKO2107",
year = "1992",
language = "English",
isbn = "951-38-4240-1",
series = "VTT Publications",
publisher = "VTT Technical Research Centre of Finland",
number = "122",
address = "Finland",
school = "University of Oulu",

}

Pehkonen, K 1992, 'Calculation of 3-D pose of a known object in a single perspective view: Dissertation', Doctor Degree, University of Oulu, Espoo.

Calculation of 3-D pose of a known object in a single perspective view : Dissertation. / Pehkonen, Kari.

Espoo : VTT Technical Research Centre of Finland, 1992. 107 p.

Research output: ThesisDissertationMonograph

TY - THES

T1 - Calculation of 3-D pose of a known object in a single perspective view

T2 - Dissertation

AU - Pehkonen, Kari

N1 - Project code: TKO2107

PY - 1992

Y1 - 1992

N2 - This thesis describes an algorithm for 3-D pose calculation of known objects from a single perspective view. The pose calculation algorithm consists of four main stages: pose candidate generation, candidate verification, optimization and reliability analysis. In the first stage, pose candidates are generated by matching model and image triangles. An important sub-problem, the so-called 3-point perspective pose estimation problem, is solved numerically. Pose candidates are verified by matching model and image features. The first accepted candidate is then optimized. Three optimization methods are considered: Newton-Raphson, simulated annealing and robust M-estimate using Lorenzian distribution. The reliability estimates are based on covariance analysis. Uncertainties due to both model and image noise are taken into account. A structure optimization method to reduce modeling noise is also presented. The algorithm was implemented on two parallel computers, a Connection Machine-2 (CM-2) and a Hathi-2/16. The Connection Machine is an SIMD machine whereas the Hathi-2/16 is based on the MIMD architecture. To evaluate the speed and accuracy of the algorithm, several experiments were performed using both synthetic and real images. In simulations, the accuracy of the calculated pose, the validity of uncertainty estimates and the effect of dynamic modeling were studied using a polyhedral test object whose maximum distance between any two vertices was 175 mm. Simulations showed that for the distance range from 0.5 m to 2.75 m, the standard deviation of the rotation error ranged from 0.5 degrees to 3.5 degrees when the standard deviation of image and model noise was 1.5 pixels and 1.5 mm, respectively. For the x-and y-directions the errors ranged from less than 2 mm up to 10 mm, and for the z-direction, from a few millimeters up to 75 mm. Much better results were achieved at the lower noise levels. The processing times on both the machines were approximately the same. On the Hathi-2/16, when all the 16 processors were used, the pose calculation time ranged from 0.376 s up to 0.814 s. For ease of programming the Hathi-2/16 was better than the CM-2, and it was also considerably cheaper.

AB - This thesis describes an algorithm for 3-D pose calculation of known objects from a single perspective view. The pose calculation algorithm consists of four main stages: pose candidate generation, candidate verification, optimization and reliability analysis. In the first stage, pose candidates are generated by matching model and image triangles. An important sub-problem, the so-called 3-point perspective pose estimation problem, is solved numerically. Pose candidates are verified by matching model and image features. The first accepted candidate is then optimized. Three optimization methods are considered: Newton-Raphson, simulated annealing and robust M-estimate using Lorenzian distribution. The reliability estimates are based on covariance analysis. Uncertainties due to both model and image noise are taken into account. A structure optimization method to reduce modeling noise is also presented. The algorithm was implemented on two parallel computers, a Connection Machine-2 (CM-2) and a Hathi-2/16. The Connection Machine is an SIMD machine whereas the Hathi-2/16 is based on the MIMD architecture. To evaluate the speed and accuracy of the algorithm, several experiments were performed using both synthetic and real images. In simulations, the accuracy of the calculated pose, the validity of uncertainty estimates and the effect of dynamic modeling were studied using a polyhedral test object whose maximum distance between any two vertices was 175 mm. Simulations showed that for the distance range from 0.5 m to 2.75 m, the standard deviation of the rotation error ranged from 0.5 degrees to 3.5 degrees when the standard deviation of image and model noise was 1.5 pixels and 1.5 mm, respectively. For the x-and y-directions the errors ranged from less than 2 mm up to 10 mm, and for the z-direction, from a few millimeters up to 75 mm. Much better results were achieved at the lower noise levels. The processing times on both the machines were approximately the same. On the Hathi-2/16, when all the 16 processors were used, the pose calculation time ranged from 0.376 s up to 0.814 s. For ease of programming the Hathi-2/16 was better than the CM-2, and it was also considerably cheaper.

KW - 3d analysis

KW - calculation

KW - parallel computing

KW - modelling

M3 - Dissertation

SN - 951-38-4240-1

T3 - VTT Publications

PB - VTT Technical Research Centre of Finland

CY - Espoo

ER -

Pehkonen K. Calculation of 3-D pose of a known object in a single perspective view: Dissertation. Espoo: VTT Technical Research Centre of Finland, 1992. 107 p.