Robert T. Collins and J. Ross Beveridge. Matching perspective views of coplanar structures using projective unwarping and similarity matching. Technical Report UM-CS-1994-006, , 1994.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....in action are given. Our interest in such planar groupings are two fold: first, once the vanishing line is obtained the affine properties of the plane are determined. The complexity of any subsequent matching operation is then reduced because the distortion is only affine rather than projective [4]; second, the grouping provides an affine invariant feature. Such a feature may be used for matching between images, for example in wide baseline stereo matching [17] or between an image and an image database, for example in image retrieval or object recognition. 2 O x X Image plane World ....

R. T. Collins and J. R. Beveridge. Matching perspective views of coplanar structures using projective unwarping and similarity matching. In Proc. CVPR, 1993.

....kind of spatial relationships between different ISR data sets is used as one module of a system for information fusion in image understanding. Keywords: Registration, Information Fusion, Image Understanding 1 Introduction Affine matching has been used mainly for image to model matching (e.g. [3, 7], see Fig. 1.a) and for image to image matching (e.g. 17, 7] see Fig. 1.b) often with the purpose of the spatial registration of images [9] Our motivation for this work is driven by the idea of a general framework of Information Fusion in Image Understanding [14, 2] To deal with multiple ....

....sets is used as one module of a system for information fusion in image understanding. Keywords: Registration, Information Fusion, Image Understanding 1 Introduction Affine matching has been used mainly for image to model matching (e.g. 3, 7] see Fig. 1. a) and for image to image matching (e.g. [17, 7], see Fig. 1.b) often with the purpose of the spatial registration of images [9] Our motivation for this work is driven by the idea of a general framework of Information Fusion in Image Understanding [14, 2] To deal with multiple visual information on all levels of abstraction requires proper ....

[Article contains additional citation context not shown here]

R.T. Collins and J.R. Beveridge. Matching perspective views of coplanar structures using projective unwarping and similarity matching. In Proc.Int.Conf. of Computer Vision and Pattern Recognition, CVPR, pages 240--245, 1993.

....projective transformation up to a particular (metric) ambiguity. This partial determination requires far less information about the world plane to be known, but is neverless sufficient to enable metric measurements of entities on the world plane to be made from their images. Collins and Beveridge [2] made a significant step in this direction by showing that once the vanishing line of the plane is identified, the transformation from world to image plane can be reduced to an affinity. They used this result to reduce the dimension of the search, from eight to six, in registering satellite ....

....angle, two equal though unknown angles, and a known length ratio. Each of these constraints can be represented simply as a circular constraint on two unknown parameters, and a closed form solution obtained by intersecting circles. This means that the problem of image registration considered in [2] can be reduced to a four dimensional search. Unlike the method of [2] no knowledge of the internal parameters of the camera is required. Faugeras et al. 7] used similar constraints in a 3D context, but only an iterative solution was given. Second, in section 3, it is shown that an imaged plane ....

[Article contains additional citation context not shown here]

R. T. Collins and J. R. Beveridge. Matching perspective views of coplanar structures using projective unwarping and similarity matching. In Proc. CVPR, 1993.

....case: ffl different source images (2D) ffl different coordinate systems. Spatial registration can be achieved by a conformal transformation (4 affine parameters) ffl the match is established at the feature level. While affine matching has been used mainly for image to model matching (e.g. [6, 12], see Fig. 1.3.a) and for image to image matching (e.g. 31, 12] see Fig. 1.3.b) this kind of matching takes place at the feature or ISR level (ISR = intermediate symbolic representation [8] This affine matching of intermediate symbolic representations with the purpose of information fusion ....

....systems. Spatial registration can be achieved by a conformal transformation (4 affine parameters) ffl the match is established at the feature level. While affine matching has been used mainly for image to model matching (e.g. 6, 12] see Fig. 1.3. a) and for image to image matching (e.g. [31, 12], see Fig. 1.3.b) this kind of matching takes place at the feature or ISR level (ISR = intermediate symbolic representation [8] This affine matching of intermediate symbolic representations with the purpose of information fusion is described in detail in [21] We focus on the situation ....

[Article contains additional citation context not shown here]

R.T. Collins and J.R. Beveridge. Matching perspective views of coplanar structures using projective unwarping and similarity matching. In Proc.Int.Conf. of Computer Vision and Pattern Recognition, CVPR, pages 240--245, 1993.

....Registration, Information Fusion, Image Understanding 1 Introduction In Computer Vision, the establishment of correspondence between different sources of visual information is an important issue. Affine matching has mainly been used for image to model matching (e.g. Beveridge et al. 1990] [Collins and Beveridge, 1993], see Fig. 1.a) and for image to image matching (e.g. Zabih and Woodfill, 1994, Collins and Beveridge, 1993, Flusser and Suk, 1994] see Fig. 1.b) often with the purpose of spatial registration of images [Brown, 1992] Our motivation for this work is driven by the idea of a general framework of ....

....of correspondence between different sources of visual information is an important issue. Affine matching has mainly been used for image to model matching (e.g. Beveridge et al. 1990] Collins and Beveridge, 1993] see Fig. 1. a) and for image to image matching (e.g. Zabih and Woodfill, 1994, Collins and Beveridge, 1993, Flusser and Suk, 1994] see Fig. 1.b) often with the purpose of spatial registration of images [Brown, 1992] Our motivation for this work is driven by the idea of a general framework of Information Fusion in Image Understanding [Pinz and Bartl, 1992a, Pinz and Bartl, 1992b, Bartl et al. ....

[Article contains additional citation context not shown here]

Collins, R. and Beveridge, J. (1993). Matching perspective views of coplanar structures using projective unwarping and similarity matching. In Proc.Int.Conf. of Computer Vision and Pattern Recognition, CVPR, pages 240-- 245.

....Our future work will extend the principle of symmetry group classi cation to patterns viewed in perspective. The rst step will be to use cues such as vanishing points to determine the horizon line of the pattern, which can be used to unwarp the image into an ane view of the surface, as shown in [3]. We will also explore applications to curved surfaces where surface geometry will play an important role. Our current emphasis is on the robustness of periodic pattern analysis algorithms to deal with the statistical nature of real images [25] ....

R.T. Collins and J.R. Beveridge. Matching perspective views of coplanar structures using projective unwarping and similarity matching. In CVPR, pages 240-245, New York City, 1993.

....of carrying out the classification procedure in perspective space directly, we take a detour into affine space. Using other cues in the image, such as two vanishing points, we construct a transformation to unwarp the projection from perspective to affine by sending the horizon line to infinity [4]. This reduces the perspective transform to an affine one, and the affine algorithm introduced above is used. If we know the camera parameters, the plane can be unwarped to a similarity transformation of the original 2D pattern [4] and we can directly apply our Euclidean algorithm from Section ....

....from perspective to affine by sending the horizon line to infinity [4] This reduces the perspective transform to an affine one, and the affine algorithm introduced above is used. If we know the camera parameters, the plane can be unwarped to a similarity transformation of the original 2D pattern [4], and we can directly apply our Euclidean algorithm from Section 3. 5 Experimental Results The examples in this section serve to illustrate the symmetry classification algorithm. The first two are synthetic images taken from a web page about wallpaper groups, maintained by David Joyce at Clark ....

R.T. Collins and J.R. Beveridge. Matching perspective views of coplanar structures using projective unwarping and similarity matching. In CVPR, pages 240--245, New York City, 1993.

....of carrying out the classification procedure in perspective space directly, we take a detour into affine space. Using other cues in the image, such as two vanishing points, we construct a transformation to unwarp the projection from perspective to affine by sending the horizon line to infinity [4]. This reduces the perspective transform to an affine one, and the affine algorithm introduced above is used. If we know the camera parameters, the plane can be unwarped to a similarity transformation of the original 2D pattern [4] and we can directly apply our Euclidean algorithm from Section 3. ....

....from perspective to affine by sending the horizon line to infinity [4] This reduces the perspective transform to an affine one, and the affine algorithm introduced above is used. If we know the camera parameters, the plane can be unwarped to a similarity transformation of the original 2D pattern [4], and we can directly apply our Euclidean algorithm from Section 3. 5 Experimental Results The examples in this section serve to illustrate the symmetry classification algorithm. The first two are synthetic images taken from a web page about wallpaper groups, maintained by David Joyce at Clark ....

R.T. Collins and J.R. Beveridge. Matching perspective views of coplanar structures using projective unwarping and similarity matching. In CVPR, pages 240--245, New York City, 1993.

.... correspondences between the two views can be found, a projective transformation mapping the second image into the coordinate frame of the first can be computed, allowing the oblique view to be effectively unwarped into registration with the initial nadir view (for an alternative approach, see [Coll93]) Figure 3 shows the final registration, as a mosaic of the extracted image line segments. One real world issue becomes immediately apparent from this example. If ideal features were being transformed between the two images, corresponding features would overlap exactly in the final registered ....

R.T. Collins and J.R. Beveridge, "Matching Perspective Views of Coplanar Structures using Projective Unwarping and Similarity Matching," Proc. IEEE Computer Vision and Pattern Recognition, New York City, June 1993, pp. 240--245. Also appeared in Proc. Darpa I.U. Workshop, Washington, DC, April 1993, pp. 459--463.

....of carrying out the classification procedure in perspective space directly, we take a detour into affine space. Using other cues in the image, such as two vanishing points, we construct a transformation to unwarp the projection from perspective to affine by sending the horizon line to infinity [4]. This reduces the perspective transform to an affine one, and the affine algorithm introduced above is used. If we know the camera parameters, the plane can be unwarped to a similarity transformation of the original 2D pattern [4] and we can directly apply our Euclidean algorithm from Section 3. ....

....from perspective to affine by sending the horizon line to infinity [4] This reduces the perspective transform to an affine one, and the affine algorithm introduced above is used. If we know the camera parameters, the plane can be unwarped to a similarity transformation of the original 2D pattern [4], and we can directly apply our Euclidean algorithm from Section 3. 5 Experimental Results The examples in this section serve to illustrate the symmetry classification algorithm. The first two are synthetic images taken from a web page about wallpaper groups, maintained by David Joyce at Clark ....

R.T. Collins and J.R. Beveridge. Matching perspective views of coplanar structures using projective unwarping and similarity matching. In CVPR, pages 240--245, New York City, 1993.

....trials of local search increase the probability of finding a near optimal solution. Our contribution has been to adapt these ideas to geometric matching problems [4] 5] 6] associated with object recognition. Our local search algorithms have been used for semiautonomous photo interpretation [7], robot navigation [8] 9] 10] and scene understanding [11] A matching system based upon the ideas presented here is now included in the KBVision system produced by AAI in Amherst Massachusetts. When estimates for object position and orientation are available, a 3D version finds optimal ....

....photographs. In these examples, models have been hand built. However, while these examples are hand built, local search has been used with real building models on the RADIUS calibrated terrain board imagery [18] It has also been used with aerial photgraphs to register ortho rectified images [7]. The data line segments for these examples are produced using the Burns algorithm [19] For the match in Figure 2d, the model is rotated by 120 ffi , illustrating that the original orientation of the model does not matter. In this example, there is little clutter and the building has a ....

Robert T. Collins and J. Ross Beveridge, "Matching perspective views of coplanar structures using projective unwarping and 15 similarity matching.," in Proceedings: 1993 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, New York, NY, June 1993, pp. 240 -- 245.

....match the outline of a truck to a subset of line segments extracted from a photograph in which a truck appears. The field of line segments may also include clutter and the matched object may be fragmented or distorted. This is a difficult problem with important practical applications [BR95, Bev93, CB93, FHR 90] It is therefore interesting to see whether genetic algorithms can improve performance beyond the baseline established using local search. Several important themes emerge from our studies. The first is the importance of distinguishing problems with strong nonlinear dependencies ....

....extracted from a photograph. Problems of this type arise in a number of areas and are specifically relevant to problems associated with computer vision. Variations of local search are currently being used to solve geometric matching problems in the context of semi autonomous photo interpretation [CB93] and robot navigation [FHR 90] In this section, we compare local search techniques to CHC and Genitor and present initial results on a set of geometric matching problems. These matching test problems are useful because factors such as problem size, object model structure, data corruption ....

Robert T. Collins and J. Ross Beveridge. Matching perspective views of coplanar structures using projective unwarping and similarity matching. In Proceedings: 1993 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pages 240 -- 245, New York, NY, June 1993.

No context found.

R. Collins, J. Beveridge, "Matching Perspective Views of Coplanar Structures using Projective Unwarping and Similarity Matching," Proc. Computer Vision and Pattern Recognition, June 1993.

No context found.

Robert T. Collins and J. Ross Beveridge. Matching perspective views of coplanar structures using projective unwarping and similarity matching. Technical Report UM-CS-1994-006, , 1994.

No context found.

Collins, R. and J. Beveridge, "Matching perspective views of coplanar structures using projective unwarping and similarity matching" Proc. IEEE Computer Vision and Pattern Recognition, 1993, pp. 240-245.

No context found.

R. Collins and J. Ross Beveridge, Matching Perspective Views of Coplanar Structures using Projective Unwarping and Similarity Matching, Proc. 1993 DARPA IUW, Washington, DC. April 19-21, 1993b.

No context found.

R.T. Collins and J.R. Beveridge. Matching perspective views of coplanar structures using projective unwarping and similarity matching. In Proc.Int.Conf. of Computer Vision and Pattern Recognition, CVPR, pages 240--245, 1993.

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