H. J. Wolfson, "Model-based object recognition by geometric hashing," in Proc. Eur. Conf. Computer Vision, Antibes, France, Apr. 1990, pp. 526--536.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....It significantly di#ers from both, however, in that it recommends a certain order of hypotheses to be tested, which is indeed the main point of the present study. Other classic concepts that are recovered naturally from the probabilistic perspective are feature based indexing and geometric hashing [13, 20], and feature grouping and perceptual organization [14, 15, 11] To keep the notational load in this article to a minimum and to support ease of reading, we will denote all probability densities by the letter p and indicate each type of density by its arguments. Moreover, we will not introduce ....

Wolfson, H. J. Model-based object recognition by geometric hashing. In Proc. Eur. Conf. Comput. Vision (1990), pp. 526--536. 16

....G(S) inf n # G # S # 1 #N, n#. 2.11) In order to decrease the time of this search, one may preprocess the data comprising the normals of Gauss Map into data structures that would minimize the query time. One can aim at organizing the Gaussian sphere into a two dimensional hash table (see [3]) possibly via a central projection, with an access to the # neighborhood of each vector on the sphere in constant time, i.e. O(1) The search of the normals which are close to a given vector on the Gaussian sphere is another operation that should be conducted e ciently. The task has been reduced ....

Wolfson HJ. Model-based object recognition by geometric hashing. Proceeding 1st European Conference on Computer Vision, Lecture Notes in Computer Science. 1990. p. 526}36.

....score measure, which leads to a more reliable recognition result. The score is traditionally calculated by counting the number of data features that are consistent with the hypothesized object instance (e.g. edge points which are close enough to the hypothesized object boundary) 9] [20]. More elaborate methods rely on an error model (usually Gaussian) implying that the contribution of every data feature to the score depends on its direction and its distance from the hypothesized boundary (e.g. 17] 13] Some recognition algorithms do not contain an explicit verification ....

H. Wolfson, "Model-Based Object Recognition by Geometric Hashing," ECCV-90, pp. 526--536, 1990.

....features of the objects, like knobs and holes, whose existence is not usually guaranteed. Other methods, which do not rely on the existence of a particular type of features, are pose clustering [78] alignment [49] and geometric hashing. A comparison between these techniques is found in [83]. Comprehensive surveys on partial surface matching techniques in computer vision are found in [11] 23] Many other works have addressed the problem, most of which have various limitations. They either restrict the shape of the matched objects (e.g. require them to be polyhedra, or to have ....

H.J. Wolfson, "Model-Based Object Recognition by Geometric Hashing," Proc. First European Conf. Computer Vision, Lecture Notes in Computer Science, 427. Berlin: Springer Verlag, 1990, pp. 526536.

....# # # #. Variations of these methods also work for geometric features other than points, such as segments, or points with normal vectors [10] and for other transformations than affine transformations. A comparison between geometric hashing, pose clustering, and the alignment method is made in [60]. Other voting schemes exist, for example taking a probabilistic approach [39] 5.2 Subdivision Schemes As mentioned above, deciding whether there is a translation plus scaling that brings the partial Hausdorff distance under a given threshold is done in [30] by a progressive subdivision of ....

H. J. Wolfson. Model-based object recognition by geometric hashing. In Proceedings of the 1st European Conference on Computer Vision, Lecture Notes in Computer Science 427, pages 526--536. Springer, 1990.

....and implementations, most of them are representative of generic paradigms for image understanding. For example, many of the following strategies fall in the category of hypothesis generation, testing, verification, and refinement. Much of this material is paraphrased from [Suetens 1992] and from [Wolfson 1990]. Finally, we show various specific areas of application in the next section. Hierarchical Models and Symbolic Constraints. One of the first object recognition systems was ACRONYM [Brooks 1983] Binford 1982] The models in ACRONYM were volumetric 3D models based on generalized cones and ....

....similarity of matching between point features to arrive at a maximum likelihood or maximum a posteriori interpretation in terms of a particular assignment (Section 3.7) Geometric Hashing and Affine Invariant Matching. Lamdan, Schwartz and Wolfson [Lamdan et al. 1988a] Wolfson and Hummel 1988] [Wolfson 1990] present a general and efficient recognition scheme using a transformation invariant hashing scheme. Invariant geometric relations among object features are used to encode modelto scene transformations using minimal feature subsets as reference coordinate frames in which other features can be ....

Wolfson, H.J. Model-Based Object Recognition by Geometric Hashing. Proceedings of the First European Conference in Computer Vision, 526---536, 1990. 142

....not very robust for errors: the success of the technique depends on the correct extraction of the graphs from the input. Another limitation of graph matching is a lack of discernment: large classes of patterns share the same graph. 10 CHAPTER 1. INTRODUCTION Geometric hashing Geometric hashing [127, 128] is another class of correspondence methods. In geometric hashing, the geometric primitives that make up a pattern are used to generate a normalised description of the pattern as a whole. For example, for finite point sets in the plane, an a#ne invariant description is generated when the point set ....

H. J. Wolfson. Model based object recognition by geometric hashing. In Proc. 1st European Conference on Computer Vision, pages 526--536, 1990.

....transformation space. The method applies to any affine group of transformations, allowing optimisations for rigid motion. Our implementation of the method performs well in terms of reliability and efficiency. 1 Introduction In applications such as pose determination [15] object recognition [26], vehicle tracking [24] optical character recognition [27] stereo matching [4] content based image retrieval [18] medical registration [25] and radiotherapy alignment [6] a major problem is finding a transformation which matches part of a pattern A to some part of another pattern B. Patterns ....

Haim J. Wolfson. Model-based object recognition by geometric hashing. First European Conference on Computer Vision, 427:526--536, 1990.

....Furthermore, the use of aspect models and more complex transformation functions in the 2 D approach causes incorrect, degenerate solutions that do not occur in the 3 D case. 1 Introduction Geometric hashing is an object recognition technique that can be used for a wide range of applications [1]. However, most papers on geometric hashing are restricted to two dimensional data. In previous papers [2, 3] we have demonstrated a combination of geometric hashing and stereo vision that uses three dimensional data. Although this use of 3 D data seems to have advantages over the 2 D case, the ....

Haim J. Wolfson. Model-based object recognition by geometric hashing. In Proceedings of the First European Conference on Computer Vision, volume 427 of Lecture Notes in Computer Vision, pages 526--536, 1990. Antibes (FR).

....features of the objects, like knobs and holes, whose existence is not usually guaranteed. Other methods, which do not rely on the existence of a certain type of features, are pose clustering [30] alignment [19] and, of course, geometric hashing. A comparison between these techniques is found in [31]. Many other works have addressed the problem; see [3, 5, 6, 8, 10, 13, 17, 18, 27] for studies in the context of object recognition, and [1, 12, 14, 20, 22, 24, 26, 28] for studies in the context of molecular biology. Most of these works have various limitations, some of which are quite severe. ....

H.J. Wolfson, Model-based object recognition by geometric hashing, Proc. 1st European Conf. on Computer Vision, Lecture Notes in Computer Science, 427, Springer Verlag, Berlin, 1990, 526--536.

....parts. Lamdan and Wolfson [10, 11] introduce the idea of geometric hashing. It uses an indexing approach based on transformation invariant representation to quickly find matched surface patches. Geometric hashing is first applied to the object recognition problem in computer vision [27]. Fischer et. al[4] and Norel et. al[16] apply the geometric hashing technique to the docking problem in molecular biology: detect the optimally matched ligand and receptor surfaces. In this paper, we will apply geometric hashing to find multi purpose fixtures. Geometric tolerancing is concerned ....

H.J. Wolfson. Model-based object recognition by geometric hashing. In Computer Vision. First European Conference on Computer Vision, pages 526--536, April 1990.

....new verification score measure, which leads to a more reliable recognition result. The score is traditionally calculated by counting the number of data features that are consistent with the hypothesized object instance (e.g. edge points which are close enough to the hypothesized object boundary) [16, 33]. More elaborate methods rely on an error model (usually Gaussian) implying that the contribution of every data feature to the score depends on its relative orientation and its distance from the hypothesized boundary (e.g. 27, 23] Some recognition algorithms do not contain an explicit ....

H. Wolfson. Model-based object recognition by geometric hashing. In ECCV-90, pages 526--536, 1990.

....The complexity of matching a single query set is O(Nm 4 n 3 ) Variations of these methods also work for geometric features other than points, and for other transformations than a ne transformations. A comparison between geometric hashing, pose clustering, and the alignment method is made in [Wol90] Other voting schemes exist, for example taking a probabilistic approach [Ols97] 2.4 Computational Geometry Computational geometry is the subarea of algorithms design that deals with geometric problems involving operations on objects like points, lines, polygons, and polyhedra. Over the past ....

H. J. Wolfson. Model-based object recognition by geometric hashing. In Proceedings of the 1st European Conference on Computer Vision, Lecture Notes in Computer Science 427, pages 526-536. Springer, 1990.

....in the appearance of the target object than is possible using a simple matched filter. A second line of research uses geometric constraints between low level object features. Methods such as alignment [11] geometric invariants [15] combinations of views [24, 21] and geometric hashing [26, 19] fit within this category. Further generalization has been obtained by allowing for an object to be represented as a collection of more complex features (or texture patches) connected with a deformable geometrical model. The neocognitron architecture [10] may be seen as an early representative. ....

H.J. Wolfson. "Model-Based Object Recognition by Geometric Hashing". In Proc. 1 st Europ. Conf. Comput. Vision, LNCS-Series Vol. 427, Springer-Verlag, pages 526--536, 1990.

....features of the objects, like knobs and holes, whose existence is not usually guaranteed. Other methods, which do not rely on the existence of a certain type of features, are poseclustering [St] alignment [HU] and, of course, geometric hashing. A comparison between these techniques is found in [Wo1]. Comprehensive surveys on partial surface matching techniques in computer vision are found in [BJ, CD] Relatively little work has been published in the area of registration (pose estimation, alignment, motion estimation) of 3 D free form shapes, and most of the existing literature addressing ....

H.J. Wolfson, Model-based object recognition by geometric hashing, Proc. 1st European Conf. on Computer Vision, 1990, 526--536.

....that a serious flaw in our system is encountered when the sub components are not rectangular in shape, with Figure 10b further demonstrating that blurring can cause ambiguities in the estimates of the rectangular shapes. To solve the problem caused by using rectangular models Geometric Hashing [7, 8, 9, 10] has been incorporated into the SOI IPA. The method uses 2 D affine point matching to estimate the orientation of the sub components. This technique assumes that: model(s) of object are known in advance; objects in a scene may overlap and or may be partially occluded; points of interest on a ....

H. Wolfson, "Model-based object recognition by geometric hashing," Computer Vision, ECCV90 , 526--536(1990).

....the largest area. This least occluded sub component is used to calculate the orientation vector of the collective sub components. Affine invariant Point Matching After the components are identified we use affine point matching to refine the 3 D pose of each component. Affineinvariant matching [7] is an appropriate technique for recognizing flat or nearly flat objects in a 2D perspective projection image. This technique assumes that: model(s) of object are known in advance; objects in a scene may overlap and or may be partially occluded; points of interest on a plane can be located; and ....

H.J. Wolfson. Model-based object recognition by geometric hashing. Computer Vision, ECCV90, pages 526--536, 1990.

....in the appearance of the target object than is possible using a simple matched filter. A second line of research has used geometric constraints between low level object features. Methods such as alignment [11] geometric invariants [15] combinations of views [24, 21] and geometric hashing [26, 19] fit within this category. Further generalization has been obtained by allowing an object to be represented as a collection of more complex features (or texture patches) connected with a deformable geometrical model. The neocognitron architecture [10] may be seen as an early representative. More ....

H.J. Wolfson. "Model-Based Object Recognition by Geometric Hashing". In Proc. 1 st Europ. Conf. Comput. Vision, LNCS-Series Vol. 427, Springer-Verlag, pages 526--536, 1990.

....Performance of the orientation estimation at varying oe 2 levels for the non rectangular solar panels of the GORZ satellite. Right: An example of how the blurring process leads to errors in the rectangular estimates. To solve the problem caused by using rectangular models Geometric Hashing [12, 6, 8, 2] has been incorporated into the SOI IPA. The method uses 2 D affine point matching to estimate the orientation of the sub components, but first we use the polygon model to replace the rectangular model. 5.1 Feature Extraction by Hough Transform The Hough transform is a technique which can be used ....

....of object are known in advance; objects in a scene may overlap and or may be partially occluded; points of interest on a plane can be located; and object depth variations in the plane are relatively small. Fortunately, SOI data satisfies these assumptions in most cases. The Hummel and Wolfson [12] affine invariant matching technique requires the use of local feature to detect interest points. The distinctiveness of points is based on sharp convexities and deep concavities along the boundary of the objects. In our case, the interest points can be the convexities or concavities of the ....

H.J. Wolfson. Model-based object recognition by geometric hashing. Computer Vision, ECCV90, pages 526--536, 1990.

.... landmarks would be to store some 3D model of a set of objects that the robot is likely to encounter and use one of the established recognition techniques such as alignment [13, 15] interpretation trees [9, 10] geometric invariance [26] aspect graphs [3, 16, 20, 27] or geometric hashing [34]. While prior models are useful for describing the destination, such an approach is going be ineffective during the course of navigation when the robot encounters many unmodelled objects. Instead, the robot should be able to learn about the new objects that it encounters and retain models of those ....

H. Wolfson. Model-based object recognition by geometric hashing. In European Conf. on Computer Vision, pages 526--536, 1990.

No context found.

H. J. Wolfson, "Model-based object recognition by geometric hashing," in Proc. Eur. Conf. Computer Vision, Antibes, France, Apr. 1990, pp. 526--536.

No context found.

H. Wolfson, Model-based object recognition by geometric hashing, in Proc. European Conf. Computer Vision, 1990, pp. 526--536.

No context found.

H. J. Wolfson, Model-based object recognition by geometric hashing, in Proceedings of the First European Conference on Computer Vision, Antibes, France, 1990, pp. 526--536.

No context found.

Wolfson, H.J., Model Based Object Recognition by 'Geometric Hashing', Proc. ECCV1, Springer-Verlag, p.526-536, 1990.

No context found.

Haim J. Wolfson. Model-based object recognition by geometric hashing. In Proc. 1st European Conference on Computer Vision, volume 427, pages 526-536. Springer, 1990.

First 50 documents

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC