Homer H. Chen. Pose determination from line-to-plane correspondences: existence condition and closed-form solutions. IEEE Transactions on Pattern Analysis and Machine Intelligence, 13(6):530--541, 1991.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....the 3 D coordinates of features through matching, are similar to range image methods [41] Detailed summaries of methods in this class are given below. Chen describes a lineto plane pose determination method in which the image data are lines that correspond to planar features on an object [42]. Point to point correspondence is not assumed. One example of its application is the projection of a light stripe across a cube generating three lines on three faces of the cube. The method shown is based on quaternions requiring the solution of an eighth degree polynomial. No initial guess is ....

H. H. Chen, "Pose determination from line-to-plane correspondences: Existence condition and closed-form solutions," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 13, no. 6, pp. 530--541, June 1991.

....of the observed point. Often using line features, rather than points, results in higher accuracy in pose estimation. This is because lines in general can be detected with higher accuracy. Closed form solutions for the estimation of camera orientation from three line correspondences are derived in [2, 1]. Navab and Faugeras [10] show, that the maximal number of solutions for the pose estimation from three skew lines is three. Due to noise, we need more than the minimum number of lines in order to obtain reasonable results. 7, 8, 9, 12] apply iterative methods and use many lines to get higher ....

....consider only line features. Tab. 1: Standard translation and rotation error E t=R in of the correct translation length (in average 26.5 length units) and the correct rotation angle (in average 55 deg. frame lies on q. Note that this symmetry consideration also holds for other methods like [12, 7, 1, 2], which do not take the endpoints of the observations into account either. Our method yields a higher precision as shown in Fig. 5. In the more general case, where one or more short segments are given, the following limit consideration is important for the robustness of the pose estimation: If ....

[Article contains additional citation context not shown here]

H.H. Chen. Pose determination from line-to-plane correspondences: existence condition and the closed-form solutions. IEEE Transactions on Pattern Analysis and Machine Intelligence, 13(6):530--541, 1991.

....Cumuli. and carefully examine their numerical stabilities under di#erent orders of substitution and elimination. For handling redundant data, iterative methods have also been developed in [12, 18, 3] Methods for camera pose from line segments instead of point features have also been developed [9, 4, 2, 13, 11]. 3 point methods intrinsically give multiple solutions. If a unique solution is required, additional information must be given. A fourth point generically su#ces, but even with an infinite number of points there are certain degenerate cases for which no unique solution is possible. These critical ....

H.H. Chen. Pose determination from line-to-plane correspondence: Existence condition and closed-form solutions. In Proceedings of the 3rd International Conference on Computer Vision, Osaka, Japan, pages 374--378, 1990.

....to the special configuration of the orthogonal trihedron. This Figure 7: A laser pointer can be considered as a second camera reduced to its optical axis. configuration appears in the image as a junction of three 2d lines. This has shown to be degenerate for the pose algorithms from lines [3, 4] since depth along the line of sight passing through the center of the trihedron is unobservable.We note u c the orientation of this line. However, one can still find from the extracted 2d lines (i.e. the h i s) the 3d orientation of each line (i.e. the u i s) using the closed form solution ....

H. Chen. Pose determination from line-to-plane correspondences: Existence solutions and closedform solutions. ieee Transactions on Pattern Analysis and Machine Intelligence, 13(6):530-- 541, June 1991.

....and for which the vertical disparity is necessarily zero. 2.2 Computing Motion by Matching Features In this section, we will examine only those methods which compute motion from point correspondences. Although algorithms using higher level features, such as lines and planes have been proposed ([22], 23] 24] 25] these usually require more than two views and also are not as practical for hardware implementation. To compute camera motion, or relative orientation, we need to find the rotation and baseline vector that best describe the transformation between the right and left camera ....

Homer H. Chen. Pose determination from line-to-plane correspondences: Existence condition and closed-form solutions. IEEE Transactions on Pattern Analysis and Machine Intelligence, 13(6):530--541, June 1991.

....of point correspondences is limited one may devise a closed form solution. Such solutions exist for 3 points (Fischler Bolles [8] 4 coplanar points (Hung al. 18] 4 points in general position (Horaud al. 16] Holt Netravali [15] and 3 lines in general position (Dhome al. 4] Chen [2]) ffl Numerical solutions. Whenever the number of point correspondences is large (greater than 6) closed form solutions are not efficient because one has to solve for a set of non linear equations where the number of equations is larger than the number of unknowns (there are six unknowns, 3 for ....

....(or equivalently each pair of point correspondences) provides two constraints which express that the object line, its corresponding image projection, and the center of projection of the camera are coplanar. This approach has already been used by Horaud al. 16] Dhome al. 4] Chen [2], Liu al. 21] and Kumar Hanson [20] The rigid transformation, whose parameters are the unknowns of the problem, is represented by a dual number quaternion. With this representation the constraints mentioned above become quadratic equations. Therefore, each line correspondence provides two ....

H. Chen. Pose determination from line-to-plane correspondences: existence solutions and closed-form solutions. IEEE Transactions on Pattern Analysis and Machine Intelligence, 13(6):530--541, June 1991.

....S i . An initial rotation component for minimization is obtained by using a geometric relationship among three segment face pairings which include intersecting line segments. In the event that the three pairings do not include intersecting line segments, a numerical polynomial based technique [3] is used to obtain a rotation component. Unfortunately, the polynomial based method is very sensitive to noise and is also computationally expensive since an eighth degree equation must be solved. On the other hand, the method which uses intersecting line segments is very fast and robust since a ....

H. H. Chen. Pose determination from line-to-plane correspondences: Existence condition and closed-form solutions. IEEE Transactions on Pattern Analysis and Machine Intelligence, 13(6):530--541, June 1991.

....full calibration is a heavy over parameterization for the pose problem, giving reduced stability and requiring more points. Mostly in computer vision, methods for pose estimation using line segments instead of points as image features have also been developed. Dhome et al. 15] and Chen [16], developed algebraic solutions for 3 line algorithms, and Lowe [17] used the NewtonRaphson method for any number of line segments. Liu, et al. 18] combined points and line segments into the same pose estimation procedure. Motivated by the lack of methods which directly provide a unique pose ....

H.H. Chen, Pose Determination from Line-to-Plane Correspondence: Existence Condition and Closed-Form Solutions,º Proc. Third Int'l Conf. Computer Vision, pp. 374±378, Osaka, Japan, 1990.

....Also using an interpretation tree search, 37] solves the pose of a polyhedra by matching a set of 3 D line segments, obtained by three light stripe range finders, to model faces; then the pose uncertainty is estimated using the covariance matrix of the endpoints of these line segments. [14] uses a polynomial approach to solve for the line to plane correspondences involved in pose determination. Based on a generalized Hough transform, 42] estimates the pose of a 3 D object by matching point triples on the object to possibly corresponding triples in the image. In the meantime, a ....

Homer H. Chen. Pose determination from line-to-plane correspondences: existence condition and closed-form solutions. IEEE Transactions on Pattern Analysis and Machine Intelligence, 13(6):530--541, 1991.

....types of sensory data, more robust object localization can be achieved. The problem of 3D object recognition 1 has received significant attention during the last two decades (e.g. see surveys [1, 2, 3, 4] Most 3D object recognition systems rely on a single type of sensory data such as vision [5, 6, 7, 8, 9, 10], range [1, 11, 12, 13] or touch [14, 15, 16] Thus, they are unsuitable for utilizing various types of sensory data which may be readily available in some tasks, such as ours, in order to improve efficiency and robustness. There have been few efforts for integrating visual and tactile data in ....

H. H. Chen. Pose determination from line-to-plane correspondences: Existence condition and closed-form solutions. IEEE Trans. on Pattern Anal. and Mach. Intell., 13(6):530--541, 1991.

....occlusions. Moreover, pose can be estimated from lines without finding their end points. Nevertheless, the perspective camera model is non linear and hence pose estimation from point and or line correspondences is non linear as well. With three line correspondences Dhome et al. 3] and Chen [1] showed that the solutions are given by an eight degree equation in one unknown. If the three lines meet at a common point, the solution is given by a fourth degree equation in one unknown [6] which degenerates to a second degree equation if the 3 D lines are mutually orthogonal. For more than ....

H. Chen. Pose determination from line-to-plane correspondences: existence solutions and closed-form solutions. IEEE Transactions on Pattern Analysis and Machine Intelligence, 13(6):530--541, June 1991.

....occlusions. Moreover, pose can be estimated from lines without finding their end points. Nevertheless, the perspective camera model is non linear and hence pose estimation from point and or line correspondences is non linear as well. With three line correspondences Dhome et al. 3] and Chen [1] showed that the solutions are given by an eight degree equation in one unknown. If the three lines meet at a common point, the solution is given by a fourth degree equation in one unknown [6] which degenerates to a second degree equation if the 3 D lines are mutually orthogonal. For more than ....

H. Chen. Pose determination from line-to-plane correspondences: existence solutions and closed-form solutions. IEEE Transactions on Pattern Analysis and Machine Intelligence, 13(6):530--541, June 1991.

....P 3 from the first view and then found the best matching triangle in the other. There are other methods which are partially related to the registration problem. For example, the one proposed by Shah and Jain [12] solved the correspondence problem in 2D by matching corners across image frames. Chen [2] has devised a technique for determining the pose of an object in a scene based on a known model. The sensory data were lines and the reference model was in the form of planes. To determine the pose of the object, a closed form solution was found for a set of line toplane correspondences. So here ....

Homer H. Chen. Pose determination from line-to-plane correspondences: Existence condition and closed-form solutions. IEEE Transactions on Pattern Analysis and Machine Intelligence, 13(6):530--541, June 1991.

....between surface orientations) and, in many cases, the object rigidity global constraint [2, 11, 12, 14, 15, 32] ffl Alignment: One or more scene feature sets of minimal size, only enough to constrain all object six degrees of freedom, are selected. Examples of these sets are three 2 D edges [7, 9], a pair of 3 D edges [30] a 2 D junction and a 3 D surface patch [6] The selected minimal scene feature sets are aligned with corresponding model ones, in order to generate a number of hypotheses about the identity of the scene object and its 3 D pose. Each generated hypothesis is then verified ....

H. H. Chen. Pose determination from line-to-plane correspondences: Existence condition and closed-form solutions. IEEE Trans. on Pattern Anal. and Mach. Intell., 13(6):530--541, 1991.

No context found.

Homer H. Chen. Pose determination from line-to-plane correspondences: existence condition and closed-form solutions. IEEE Transactions on Pattern Analysis and Machine Intelligence, 13(6):530--541, 1991.

No context found.

CHEN H. H.: Pose determination from line-toplane correspondences: Existence condition and closed-form solutions. IEEE Transactions on Pattern Analysis and Machine Intelligence 13, 6 (1991), 530--541.

No context found.

CHEN H. H.: Pose determination from line-toplane correspondences: Existence condition and closed-form solutions. IEEE Transactions on Pattern Analysis and Machine Intelligence 13, 6 (1991), 530--541.

No context found.

H. Chen. Pose determination from line-to-plane correspondences: Existence solutions and closedform solutions. Transactions on Pattern Analysis and Machine Intelligence, 13(6):530-541, June 1991.

No context found.

H.H. Chen, Pose determination from line-to-plane correspondences: existence condition and closed-form solutions, IEEE Trans. Pattern Anal. Mach. Intell. 13 (6) (1991) 530}541.

No context found.

Chen, H. H. 1991. Pose determination from line-to-plane correspondences: existence condition and closed-form solutions. IEEE Trans. Pattern Anal. Machine Intell. 13(6):530--541.

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