Model-based 3D object detection

Biegelbauer, G.; Vincze, Markus; Wohlkinger, Walter

doi:10.1007/s00138-008-0178-3

Cited by 21 publications

(16 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We can consider a relationship between a hammer handle and how far it is from the head, and at what angle (however this is not the only paper to consider multiple parts, e.g. see [7], [9], [10], [11]).…”

Section: Related Workmentioning

confidence: 99%

“…The approach of Biegelbauer et al [7] was the starting point for the construction of our system. Biegelbauer and Vincze's system is able to quickly fit geometric models of common shapes (they use superquadrics, here explained in Section III) onto point cloud data.…”

Section: Related Workmentioning

confidence: 99%

“…Our system has three essential components: (i) Handcoded models of ideal objects for tasks; this is similar to several works in this area defining how objects are composed from parts [6], [7], [8]. (ii) Task models that specify the relative importance of various parameters of the model for a particular task.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A model-based approach to finding substitute tools in 3D vision data

Abelha

Guérin

Schoeler

2016

2016 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

A robot can feasibly be given knowledge of a set of tools for manipulation activities (e.g. hammer, knife, spatula). If the robot then operates outside a closed environment it is likely to face situations where the tool it knows is not available, but alternative unknown tools are present. We tackle the problem of finding the best substitute tool based solely on 3D vision data. Our approach has simple hand-coded models of known tools in terms of superquadrics and relationships among them. Our system attempts to fit these models to point clouds of unknown tools, producing a numeric value for how good a fit is. This value can be used to rate candidate substitutes. We explicitly control how closely each part of a tool must match our model, under direction from parameters of a target task. We allow bottom-up information from segmentation to dictate the sizes that should be considered for various parts of the tool. These ideas allow for a flexible matching so that tools may be superficially quite different, but similar in the way that matters. We evaluate our system's ratings relative to other approaches and relative to human performance in the same task. This is an approach to knowledge transfer, via a suitable representation and reasoning engine, and we discuss how this could be extended to transfer in planning.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

A model-based approach to finding substitute tools in 3D vision data

Abelha

Guérin

Schoeler

2016

2016 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

show abstract

“…The work by Tarbox and Gottschlich () is an example where the visibility of points on an imaging sphere to those on the object model is used to derive the optimum locations and directions of the scanner. Other techniques (for example, Ellenrieder et al., ; Biegelbauer et al., ) use a similar concept where a certain optimisation model involving the object and viewing points is used to identify the optimum scanner positions and orientations within the allowed viewpoint workspace.…”

Section: Literature Reviewmentioning

confidence: 99%

Optimal Placement of a Terrestrial Laser Scanner with an Emphasis on Reducing Occlusions

Heidarimozaffar

Varshosaz

2016

The Photogrammetric Record

View full text Add to dashboard Cite

In this paper, a new automated algorithm is proposed that finds the optimum locations of a terrestrial laser scanner (TLS), ensuring completeness of data and minimising the number of scanning locations. The process starts with an initial scan and placing a 3D grid of candidate stations over the entire scan area. A global visibility analysis is then performed to identify the next best view (NBV) location. The TLS is placed on this selected point and a new scan is recorded. Having updated the initial scan with the resulting point cloud, the model is checked for completeness and density. The process is repeated until full coverage of the scan area is achieved by determining the best global arrangement with the minimum number of stations. Experiments show that the algorithm is able to automatically determine the station positions and provide a coverage of 99Á5% for simulated data and 91% for real data.

show abstract

“…We assume that only geometrical attributes (shape and pose) of the objects in addition to the point cloud of the target environment are available to our algorithms. Such data, for example, can be obtained by utilizing object detection and pose estimation methods [4][5][6][7].…”

Section: Introductionmentioning

confidence: 99%