Uncertainty-driven viewpoint planning for 3D object measurements

Li, Y.F.; Liu, Z.G.

doi:10.1109/robot.2003.1241584

Cited by 14 publications

(7 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Equation (2) can be expressed as a linear combination of the control points in the B-spline representation,…”

Section: B-spline Approximationmentioning

confidence: 99%

See 1 more Smart Citation

Untitled

Dianov

2007

PHYS-USP

View full text Add to dashboard Cite

In this paper we present a novel method for determining the probing points for achieving efficient measurement and reconstruction of freeform surfaces. A B-spline is adopted for modelling the freeform surface. In the framework of Bayesian statistics, we develop a model selection strategy to obtain an optimal model structure for the freeform surface. Based on the selected model structure, a set of probing points is then determined where measurements are to be made. In order to obtain reliable parameter estimation for the B-spline model, we analyse the uncertainty of the model and use the statistical analysis of the Fisher information matrix to optimize the locations of the probing points needed in the measurements. Using a 'data cloud' of a surface acquired by a 3D vision system, we implemented the proposed method for reconstructing freeform surfaces. The experimental results show that the method is effective and promises useful applications in multi-sensor measurements including a vision guided coordinate measuring machine for reverse engineering.

show abstract

“…Equation (2) can be expressed as a linear combination of the control points in the B-spline representation,…”

Section: B-spline Approximationmentioning

confidence: 99%

“…Reconstructing the freeform surface from a set of discrete measurement data points is a problem important to many areas including reverse engineering, metrology, inspection by machine vision and computer aided design [1][2][3][4][5]. The first task in the reconstruction of a freeform surface is to obtain the measurement data.…”

Section: Introductionmentioning

confidence: 99%

Untitled

Dianov

2007

PHYS-USP

View full text Add to dashboard Cite

show abstract

“…The term "best-next-view" (BNV) was defined as the next sensor pose which would enable the greatest amount of previously unseen three-dimensional information to be acquired [13,15]. [3] [34] chose to formulate the probing strategy as a function minimization problem.…”

Section: A Previous Approachesmentioning

confidence: 99%

“…Comparing with an octree representation of a 3D scene [12], Banta and Abidi used uniformly sized voxels to represent the viewing volume [13]. The next best view (NBV) [14,15] was identified as the one that would sample the most nonempty voxels and achieve maximum information gain in the object model.…”

Section: Introductionmentioning

confidence: 99%

Automatic Sensor Placement for Model-Based Robot Vision

Chen

2004

IEEE Trans. Syst., Man, Cybern. B

Self Cite

161

View full text Add to dashboard Cite

Abstract-This paper presents a method for automatic sensor placement for model-based robot vision. In such a vision system, the sensor often needs to be moved from one pose to another around the object to observe all features of interest. This allows multiple 3D images to be taken from different vantage viewpoints. The task involves determination of the optimal sensor placements and a shortest path through these viewpoints. During the sensor planning, object features are resampled as individual points attached with surface normals. The optimal sensor placement graph is achieved by a genetic algorithm in which a min-max criterion is used for the evaluation. A shortest path is determined by Christofides algorithm.A Viewpoint Planner is developed to generate the sensor placement plan. It includes many functions, such as 3D animation of the object geometry, sensor specification, initialization of the viewpoint number and their distribution, viewpoint evolution, shortest path computation, scene simulation of a specific viewpoint, parameter amendment. Experiments are also carried out on a real robot vision system to demonstrate the effectiveness of the proposed method.

show abstract

“…The work by Pito [8] removed the need to ray-cast from every possible sensor location by determining a subset of positions that would improve the current model. An uncertainty driven approach was investigated [9], to maximize the information gain for the next view.…”

Section: Introductionmentioning

confidence: 99%

Vision Sensor Planning for 3–D Model Acquisition

Chen

2005

IEEE Trans. Syst., Man, Cybern. B

Self Cite

109

View full text Add to dashboard Cite

Abstract-A novel method is proposed in this paper for automatic acquisition of three-dimensional models of unknown objects by an active vision system, in which the vision sensor is to be moved from one viewpoint to the next around the target to obtain its complete model. In each step, sensing parameters are determined automatically for incrementally building the 3D target models. The method is developed by analyzing the target's trend surface, which is the regional feature of a surface for describing the global tendency of change. Whilst previous approaches to trend analysis are usually focused on generating polynomial equations for interpreting regression surfaces in three dimensions, this paper proposes a new mathematical model for predicting the unknown area of the object surface. A uniform surface model is established by analyzing the surface curvatures. Furthermore, a criterion is defined to determine the exploration direction and an algorithm is developed for determining the parameters of the next view. Implementation of the method is carried out to validate the proposed method.Index Terms-vision sensor, viewpoint planning, sensor placement, 3D modeling, trend surface, surface prediction, model acquisition.

show abstract

Uncertainty-driven viewpoint planning for 3D object measurements

Cited by 14 publications

References 16 publications

Untitled

Untitled

Automatic Sensor Placement for Model-Based Robot Vision

Vision Sensor Planning for 3–D Model Acquisition

Contact Info

Product

Resources

About