D. J. Kreigman and J. Ponce, "On recognizing and positioning curved 3D objects from image contours," IEEE Trans. Pattern Anal. Machine Intell., vol. 12, pp. 1127--1137, Dec. 1990.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....we implemented in our system. Two kinds of features will be tracked: points, used to determine the viewpoint, and curve segments (the arches) The arches are not used directly as curved segments for viewpoint determination in the present version of the system, as could be done with the method of [16], but only because they provide interesting feature points (the top of each arch for instance see point 4 of Fig. 11.c) In fact, we will see that even points are tracked using a curve based tracking tool. Features to be tracked are selected before execution in the first image and the ....

....inversion of curved segments. If curved arcs are present in the scene, they may be used to further converge to the position of the viewpoint. Except in the case of circles [11] an exact perspective inversion for ellipses is not possible. We may thus use an iterative algorithm developed by [16], taking as initial estimate the perspective projection matrix obtained for a number of (coplanar or non coplanar) 18 a. 86.5 86 85.5 85 84.5 b. 10.57 10.572 10.574 10.576 10.578 10.58 0 50 100 150 200 250 300 c. 317.1 317.2 317.3 317.4 317.5 317.6 317.7 0 50 100 150 ....

Kriegman, D. and Ponce, J. (1990). On recognizing and positioning curved 3D objects from image contours. IEEE Transactions on Pattern Analysis and Machine Intelligence, 12(12):1127--1137.

....iterative algorithm that does not require initial estimates and performs in realtime [DEME95] However, their method uses scaled orthographic projections and did not fully used the fact that rotation matrices are orthonormal. Researchers also used curves or surfaces for pose estimation [FELD97] KRIE90] which are not relevant to our problem. Recently, researchers from LORIA combined 3D features and 2D correspondences to compute accurate camera pose changes [SIMO98] Because the relationship between corresponding points is a function of camera motion, the camera pose accuracy can be improved ....

D. Kriegman and J. Ponce, "On Recognizing and Positioning Curved 3D Objects from Image Contours", IEEE Transactions on PAMI, 12(12):1127-137, December 1990

....recognition is the domain where a model exists for every object in the recognition system s universe of discourse. The research emphasis in this paradigm has historically been on the design of efficient matching algorithms from a manually designed feature set with hand crafted shape rules [2] 3] [4] [5] Manually designing a feature set is appealing because such a feature set is very efficient. When designed properly, a very small number of parameters for each of the objects is sufficient to capture the distinguishing characteristics among the objects to be recognized. This pre meditated ....

D. J. Kriegman and J. Ponce, "On recognizing and positioning curved 3-D objects from image contours," IEEE Trans. Pattern Anal. Machine Intell., vol. 12, no. 12, pp. 1127--1137, 1990.

....describes an iterative least squares algorithm for aligning the projected extreme contours of the model with edges found in the image. However, this technique assumes a polyhedral model; and would be difficult to apply to the knee implants, which have smooth, complex surfaces. Kriegman and Ponce [23] used rational surface patches, implicit algebraic equations, and elimination theory to obtain analytic expressions for the projected contours. However, this method is restricted to objects with only a few patches, and would be difficult to apply to knee components, which have highly complex ....

D. J. Kriegman and J. Ponce, "On Recognizing and Positioning Curved 3D Objects From Image Contours," IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. PAMI-12, No. 12, pp. 1127-1137, 1990.

....Implicit algebraic curves and surfaces (i.e. implicit 2D polynomial curves and 3D polynomial surfaces) are very attractive for modeling 2D and 3D shapes. There was great excitement in certain circles concerning implicit polynomials for use in computer vision during the late 80 s and early 90 s [1, 2, 3, 4, 5, 6, 7] because they were mathematically interesting, had not been considered previously, were potentially powerful generalizations of the conics, had the nice property of being implicit so that fitting was conceptually uncomplicated it was not necessary to deal with inconvenient view dependent ....

D. Kriegman and J. Ponce, "On recognizing and positioning curved 3D objects from image contours," IEEE PAMI, December 1990.

....These approaches use only a limited amount of the structural information available, however. Another approach to recognizing smooth 3 D objects involves describing the 3 D object and 2 D image with algebraic surfaces and curves, and then registering these algebraic descriptions. Kriegman and Ponce[31] have taken this approach, using elimination methods to solve for object pose. While this approach has provided significant insight into how the overall problem may be solved, it has the disadvantage of requiring a somewhat complex, iterative solution method. Specifically, their method requires a ....

Kriegman, D. and J. Ponce, 1990, "On Recognizing and Positioning Curved 3-D Objects from Image Contours," IEEE Trans. Pattern Anal. Machine Intell., 12(12): 1127--1137.

....to reliably locate local features in images of 3 D objects. Second, local methods can be computationally expensive, requiring extensive search to find correspondences. Techniques have been applied to the recognition of curved objects that use algebraic descriptions of extended curves (e.g. [18, 23, 11, 17]) This approach can potentially overcome some of the difficulties of isolating local features reliably. However, the need for extended contour segments can make this approach vulnerable to partial occlusion, and it can be difficult to robustly extract an algebraic description of a curved contour ....

Kriegman, D. and J. Ponce, 1990, "On Recognizing and Positioning Curved 3-D Objects from Image Contours," IEEE Trans. on PAMI, 12(12): 1127--1137.

....a point source (F) pass through the object (such as point G) and strike the image plane (point Q) 6 with edges found in the image. However, this technique assumes a polyhedral model; and would be difficult to apply to the knee implants, which have smooth, complex surfaces. Kriegman and Ponce [31] used rational surface patches, implicit algebraic equations, and elimination theory to obtain analytic expressions for the projected contours. However, this method is restricted to objects with only a few patches, and would be difficult to apply to knee components, which have highly complex ....

D. J. Kriegman and J. Ponce, "On Recognizing and Positioning Curved 3D Objects From Image Contours," IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. PAMI-12, No. 12, pp. 11271137, 1990.

....approaches use only a limited amount of the structural information available, however. Another approach to recognizing smooth 3 D objects involves describing the 3 D object and 2 D image with algebraic surfaces and curves, and then registering these algebraic descriptions. Kriegman and Ponce [31] have taken this approach, using elimination methods to solve for 1 object pose. While this approach has provided significant insight into how the overall problem may be solved, it has the disadvantage of requiring a somewhat complex, iterative solution method. Specifically, their method requires ....

D. Kriegman and J. Ponce, 1990, "On Recognizing and Positioning Curved 3-D Objects from Image Contours," IEEE Trans. on Pattern Analysis Machine Intelligence, 12(12): 1127--1137.

.... alignment is tested by searching for other model to image correspondence predicted by the model pose (verification) This algorithm has been implemented for a variety of data formats and feature types [1, 6, 15, 24, 38, 69] In fact, extensions to 3D curved surfaces have even been created [13, 33]. 3. Pose clustering is implemented by computing the object pose from a group of features corresponding to a particular model, and storing the estimate in an accumulator in pose space; if enough local groups have the same pose, a hypothesis for the model is formed. This approach (frequently called ....

Kriegman, D.J. and Ponce, J. "On Recognizing and Positioning Curved 3-D Objects from Image Contours," IEEE Trans. PAMI, Vol. 12, No. 12, p.1127-1137, December 1990.

....much worse parameter estimates in the presence of noisy data. Therefore, he adopts a similar simultaneous minimization as is used in the work above. A quite different approach based on the use of elimination methods to provide the initial problem formulation has been proposed by Ponce and Kriegman [29]. This also uses Newton s method for the final parameter determination based on least squares minimization. Haralick et al. 11] have experimented with robust methods such as iterative reweighting in order to allow for outliers caused by incorrect matches. However, their results show that even one ....

Ponce, Jean, and David J. Kriegman, "On recognizing and positioning curved 3D objects from image contours," DARPA Image Understanding Workshop, Palo Alto, CA (1989), 461--470.

....of the database. As such, content based image retrieval is fundamentally an object recognition problem. The research emphasis to this end has historically been on the design of efficient matching algorithms from a manually designed feature set with hand crafted shape rules [2] 6] 10] 11] 12] [17] [18] 28] Handcrafted shape rules can exploit the efficiency found in manually tuning features for a particular training image set. However, these rules have a severe limitation on the type of object classes that can be found by the image retrieval system. Objects greatly different than those ....

D. J. Kriegman and J. Ponce, "On recognizing and positioning curved 3-D objects from image contours," IEEE Trans. Pattern Anal. Machine Intell., vol. 12, no. 12, pp. 1127--1137, 1990.

....Two kinds of features will be tracked: points, used to determine the viewpoint, and curve segments (the arches) The arches are not used directly as curved segments for viewpoint determination in 18 M. O. Berger et al. the present version of the system, as could be done with the method of (Kriegman and Ponce, 1990), but only because they provide interesting feature points (the top of each arch for instance see point 4 of Fig. 11.c) In fact, we will see that even points are tracked using a curvebased tracking tool. Features to be tracked are selected before execution in the rst image and the ....

....of curved segments. If curved arcs are present in the scene, they may be used to further converge to the position of the viewpoint. Except in the case of circles (Ferri et al. 1993) an exact perspective inversion for ellipses is not possible. We may thus use an iterative algorithm developed by (Kriegman and Ponce, 1990), taking as initial estimate the perspective projection matrix obtained for a number of (coplanar or non coplanar) points. We did not use this feature in the application described because the modelling of the arches was too inconsistent. Smoothing the sequence. We do not currently ensure any ....

Kriegman, D. and Ponce, J. (1990). On recognizing and positioning curved 3D objects from image contours. Ieee Transactions on Pattern Analysis and Machine Intelligence, 12(12):1127{ 1137.

....found (m 3 n 3 matches) Techniques have been applied to the recognition of curved objects that are intermediate between the local and global methods. These recognition methods match algebraic descriptions of the surfaces of objects to the curves found in images. This is done, for example, by [27, 43, 16, 25]. These methods rely on locating extended portions of an object s contour, but not the entire contour. This approach can potentially overcome some of the difficulties of isolating local features reliably. Also, matching a single curve fragment may in principle provide enough information to ....

Kriegman, D. and J. Ponce, 1990, "On Recognizing and Positioning Curved 3-D Objects from Image Contours," IEEE Trans. Pattern Anal. Machine Intell., 12(12): 1127--1137.

No context found.

D. J. Kreigman and J. Ponce, "On recognizing and positioning curved 3D objects from image contours," IEEE Trans. Pattern Anal. Machine Intell., vol. 12, pp. 1127--1137, Dec. 1990.

No context found.

# D.J. Kriegman and J. Ponce, "On Recognizing and Positioning Curved 3D Objects from Image Contours," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 12, no. 12, pp. 1,127-1,138, Dec. 1990.

No context found.

D.J. Kriegman and J. Ponce, "On recognizing and positioning curved 3D objects from image contours ", IEEE Transactions On Pattern Analysis and Machine Intelligence, 12(12), pp. 1127--1137 (December 1990).

No context found.

D.J. Kriegman and J. Ponce, "On recognizing and positioning curved 3D objects from image contours," IEEETrans. on Pattern Analysis and Machine Intelligence, Vol. 12, 1127--1138, 1990.

No context found.

D. Kriegman and J. Ponce, "On recognizing and positioning curved 3-D objects from image contours ", IEEE Trans. on Pattern Anal. and Machine Intell., PAMI 12, no. 12, pp 1127-1137,1990

No context found.

D.J. Kriegman and J. Ponce, "On Recognizing and Positioning Curved 3D Objects from Image Contours," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 12, no. 12, pp. 1127-1138, Dec. 1990.

No context found.

# D.J. Kriegman and J. Ponce, "On Recognizing and Positioning Curved 3D Objects from Image Contours," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 12, no. 12, pp. 1,127-1,138, Dec. 1990.

No context found.

D. Kriegman and J. Ponce, "On recognizing and positioning curved 3-D objects from image contours", IEEE Trans. on Pattern Anal. and Machine Intell., PAMI 12, no. 12, pp 1127-1137,1990

No context found.

D. J. Kriegman and J. Ponce, "On recognizing and positioning curved 3-D objects from image contours," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 12, no. 12, pp. 1127--1137, 1990.

No context found.

D. J. Kriegman and J. Ponce, "On recognizing and positioning curved 3-d objects from image contours," IEEE Trans. Pattern Anal. Machine Intell., vol. 12, no. 12, pp. 1127--1137, 1990.

No context found.

Kriegman, D.J., and J. Ponce, "On Recognizing and Positioning Curved 3-D Objects from Image Contours," IEEE Trans. on PAMI-12, pp. 1127-1137, 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