## Robust and fast modelling of 3D natural objects from multiple views (1994)

Venue: | SPIE Proceedings "Image and Video Processing II |

Citations: | 33 - 5 self |

### BibTeX

@INPROCEEDINGS{Niem94robustand,

author = {Wolfgang Niem and Universität Hannover},

title = {Robust and fast modelling of 3D natural objects from multiple views},

booktitle = {SPIE Proceedings "Image and Video Processing II},

year = {1994},

pages = {388--397}

}

### Years of Citing Articles

### OpenURL

### Abstract

An algorithm for the robust and fast automatic construction of a 3D model of any real object using images from multiple views is presented. The images are taken from a real object rotating in front of a stationary calibrated CCD TV camera. The object silhouettes extracted from the input images, the related turntable positions and camera orientation are used to construct the volume model of the real object by applying the method of occluding contours. A keypoint in performing this method is a proper volume representation, characterized by low complexity and suitability for a fast computation of volume models. In the presented approach, each volume model is described by pillar–like volume elements (pillars) ensuring a computational complexity proportional to the size of the real object surface and enabling a fast and simple construction of the volume model. The fast performance is due to the simple projection feasibility of those pillars and the easy–to–perform intersection test for the object silhouette with the projected pillars. Results with real image sequences have confirmed the robustness of the developed algorithm even for the modelling of real objects with highly detailed and complex surfaces and the use of imperfect object silhouettes. 1

### Citations

1194 |
A Versatile Camera Calibration Technique for High-Accuracy 3D Machine Vision Metrology Using Off-the-Shelf TV Cameras and Lenses
- Tsai
- 1987
(Show Context)
Citation Context ... � � � � � � � � ��� � � � � � � � �� � � � � � � � � The main advantage of the camera description with the � � � � � � � � � � � system in contrary to other alternatives like the system used by Tsai =-=[14]-=- is the easy–to–perform projection of a real world point into the image plane of a CCD camera. All geometric manipulations described in the next sections are performed using this � � � � � � � � � � �... |

693 | A computer algorithm for reconstructing a scene from two projections. Nature 293 - Longuet-Higgins - 1981 |

242 |
Rapid octree construction from image sequences
- Szeliski
- 1993
(Show Context)
Citation Context ...ation, the octree representation, which describes the object as a tree of recursively subdivided cubes. Chien [3] constructed an octree from three orthographic projections. Potmesil [11] and Szeliski =-=[13]-=- used arbitrary views and perspective projection to construct an octree. Both of these approaches project the octree cubes into the image plane to perform the intersection test between the projected c... |

124 |
Geometric modeling using octree encoding
- Meagher
- 1982
(Show Context)
Citation Context ...a volume model. Busch [1] used a polygon approximation of the object silhouette and constructed a volume model from arbitrary views using a voxel representation. Jackins and Tanimoto [5], and Meagher =-=[9]-=- introduced an efficient volume representation, the octree representation, which describes the object as a tree of recursively subdivided cubes. Chien [3] constructed an octree from three orthographic... |

120 |
Volumetric descriptions of objects from multiple views
- Martin, Aggarwal
- 1983
(Show Context)
Citation Context ... real objects from multiple views presented in this paper makes a contribution to the passive methods. Basic work for the construction of 3D models from multiple views was done by Martin and Aggarwal =-=[8]-=-. They introduced the method of ”occluding contours” which uses the real object silhouettes extracted from the real images and orthographic projection for the construction of a volume model. Busch [1]... |

93 |
Generating octree models of 3d objects from their silhouettes in a sequence of images
- Potmesil
- 1987
(Show Context)
Citation Context ...t volume representation, the octree representation, which describes the object as a tree of recursively subdivided cubes. Chien [3] constructed an octree from three orthographic projections. Potmesil =-=[11]-=- and Szeliski [13] used arbitrary views and perspective projection to construct an octree. Both of these approaches project the octree cubes into the image plane to perform the intersection test betwe... |

67 |
et al., Computer Graphics: Principles & Practice, 2nd edition
- Foley
- 1996
(Show Context)
Citation Context ...� � � � � � � � � � , i = 1, 2 (9) � � � ��� � � � � � � � � �� � � � � � � � � � , i = 1, 2 (10) Connecting the resulting image points with a 2D line (Fig. 6 b) A scan–conversion algorithm for lines =-=[4]-=- computes the coordinates of the pixels that lie on or near an ideal, infinitely thin straight line through � � � , � � � imposed on a 2D raster grid. This line represents the projection of a volume p... |

61 |
Oct-trees and their use in representing threedimensional objects
- Jackins, Tanimoto
- 1980
(Show Context)
Citation Context ... construction of a volume model. Busch [1] used a polygon approximation of the object silhouette and constructed a volume model from arbitrary views using a voxel representation. Jackins and Tanimoto =-=[5]-=-, and Meagher [9] introduced an efficient volume representation, the octree representation, which describes the object as a tree of recursively subdivided cubes. Chien [3] constructed an octree from t... |

46 |
A system for extracting three-dimensional measurement forms a stereo pair of TV cameras
- Yakimovsky, Cunninggham
- 1978
(Show Context)
Citation Context ...l object surface and enabling a fast and simple intersection test of the projected volume elements with the object silhouette. The camera model used is the one introduced by Yakimovski and Cunningham =-=[15]-=- which is suited for a fast implementation of the central projection. In Section 2 the camera model used for the modelling is defined. Section 3 describes the measurement environment and the system ca... |

26 |
Identification of 3d objects from multiple silhouettes using quadtrees/octrees
- Chien, Aggarwal
- 1986
(Show Context)
Citation Context ...tation. Jackins and Tanimoto [5], and Meagher [9] introduced an efficient volume representation, the octree representation, which describes the object as a tree of recursively subdivided cubes. Chien =-=[3]-=- constructed an octree from three orthographic projections. Potmesil [11] and Szeliski [13] used arbitrary views and perspective projection to construct an octree. Both of these approaches project the... |

22 |
A survey of construction and manipulation of octrees
- Chen, Huang
- 1988
(Show Context)
Citation Context ...he method of ”occluding contours” is a proper volume representation, characterized by low complexity and suitability for a fast computation of volume models. One often used representation are octrees =-=[2]-=-,[3],[5],[6],[9],[10],[11],[13], which describe the object as a tree of recursively subdivided cubes down to the finest resolution. The volume model is constructed projecting the octree cubes consecut... |

19 |
Construction of the octree approximating three dimensional objects by using multiple views
- Noborio, Fukuda, et al.
- 1988
(Show Context)
Citation Context ...which can lead to a time consuming intersection test. The approximation of this hexagon by a bounding box as done by several authors can lead to reconstruction errors. Moreover, some authors [1],[12],=-=[10]-=- approximate the object silhouette with a polygon, which can lead to reconstruction errors as well, especially in case of complexssilhouettes with holes. The speed of computation also depends on the w... |

17 | Octree generation from object silhouettes in perspective views
- Srivastava, Ahuja
- 1990
(Show Context)
Citation Context ...agon which can lead to a time consuming intersection test. The approximation of this hexagon by a bounding box as done by several authors can lead to reconstruction errors. Moreover, some authors [1],=-=[12]-=-,[10] approximate the object silhouette with a polygon, which can lead to reconstruction errors as well, especially in case of complexssilhouettes with holes. The speed of computation also depends on ... |

8 | Longuet--Higgins, "A computer algorithm for reconstructing a scene from two projections - C - 1981 |

5 |
et al., "Computer Graphics: Principles and Practice
- Foley
- 1992
(Show Context)
Citation Context ... # ## # # # # # # # # , i = 1, 2 (9) # # ## # # # # # # # # ## # # # # # # # # # , i = 1, 2 (10) Connecting the resulting image points with a 2D line (Fig. 6 b) A scan--conversion algorithm for lines =-=[4]-=- computes the coordinates of the pixels that lie on or near an ideal, infinitely thin straight line through # # # , # # # imposed on a 2D raster grid. This line represents the projection of a volume p... |

4 |
Oct-Trees and Their Use
- Jackins, Tanimoto
- 1980
(Show Context)
Citation Context ... construction of a volume model. Busch [1] used a polygon approximation of the object silhouette and constructed a volume model from arbitrary views using a voxel representation. Jackins and Tanimoto =-=[5]-=-, and Meagher [9] introduced an efficient volume representation, the octree representation, which describes the object as a tree of recursively subdivided cubes. Chien [3] constructed an octree from t... |

4 |
Ein Verfahren zur Oberfl��chenmodellierung dreidimensionaler Objekte aus Kamerabildfolgen
- Busch
- 1991
(Show Context)
Citation Context ...[8]. They introduced the method of "occluding contours" which uses the real object silhouettes extracted from the real images and orthographic projection for the construction of a volume mod=-=el. Busch [1]-=- used a polygon approximation of the object silhouette and constructed a volume model from arbitrary views using a voxel representation. Jackins and Tanimoto [5], and Meagher [9] introduced an efficie... |

3 |
Linear octrees by volume intersection
- Lavakusha, Reddy
- 1989
(Show Context)
Citation Context ... ”occluding contours” is a proper volume representation, characterized by low complexity and suitability for a fast computation of volume models. One often used representation are octrees [2],[3],[5],=-=[6]-=-,[9],[10],[11],[13], which describe the object as a tree of recursively subdivided cubes down to the finest resolution. The volume model is constructed projecting the octree cubes consecutively into t... |

2 |
Aggarwal , "Volumetric Descriptions of Objects from Multiple Views
- Martin, K
- 1983
(Show Context)
Citation Context ... real objects from multiple views presented in this paper makes a contribution to the passive methods. Basic work for the construction of 3D models from multiple views was done by Martin and Aggarwal =-=[8]. They int-=-roduced the method of "occluding contours" which uses the real object silhouettes extracted from the real images and orthographic projection for the construction of a volume model. Busch [1]... |

1 |
Ein Verfahren zur Oberflächenmodellierung dreidimensionaler Objekte aus Kamerabildfolgen
- Busch
- 1991
(Show Context)
Citation Context ...[8]. They introduced the method of ”occluding contours” which uses the real object silhouettes extracted from the real images and orthographic projection for the construction of a volume model. Busch =-=[1]-=- used a polygon approximation of the object silhouette and constructed a volume model from arbitrary views using a voxel representation. Jackins and Tanimoto [5], and Meagher [9] introduced an efficie... |