Dreschler L., Nagel H.H.; Volumetric model and 3D trajectory of a moving car derived from monocular TV frame sequences of a street scene; Proc. Int. Joint Conf. Artificial Intelligence, 1981, p.692-697.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....world surrounding the system is not considered. Schwind 4 characterised this approach as follows: A semantic concept The work described in this paper was supported by the German Special Collaboration Project SFB 314 on AI and Knowledge Based Systems of the German Science Foundation (DFG) cf. Dreschler, Nagel 82] 2 cf. Nebel, Marburger 82] and [Marburger, Wahlster 83] 3 cf. Neumann, Novak 86] 4 cf. Schwind 84] p. 96; shall simulate or represent something, which we do not know directly: conceptual structures, that humans construct to understand and produce sentences. With respect to ....

L. Dreschler, H.-H. Nagel (1982): Volumetric Model and 3D-Trajectory of a Moving Car Derived from Monocular TV-Frame Sequences of a Street Scene. In: Computer Graphics and Image Processing, Vol. 20, pp. 199-228.

....interest due to their accurate localisation. Sensitivity to noise, speed and selectivity are all issues which need to be considered when attempting to implement a corner detector. A number of methods for detecting corners within shapes include template matching, second order derivative schemes [14, 15], autocorrelation [16] median based methods [17] and the interesting point operator [18] The corner detector implemented in the following experiments is particularly suited to detecting corners which are projections of vertices. It locates corners by first finding regions where several lines ....

L.Dreschler and H.-H.Nagel. Volumetric model and 3d-trajectory of a moving car derived from monocular tv-frame sequences of a street scene. Proc. It. Joint Cof. o Artif. Itell., pages 692 697, 1981.

....Comparisons of the method in different scales and with the SUSAN corner detector are also given. 1 Introduction Corners are important dominant points in digital images. In many computer vision tasks, such as image registration, image matching [4] object recognition [15, 10] and motion analysis[7], accurate corner detection is essential. Broadly speaking, there are two cornerdetection strategies adopted in literature. The first of these is based on the analysis of pre segmented contours, while the second is based on the differential analysis of the raw gray scale image. However, in both ....

L. Dreschler and H. Nigel. Volumetric model and 3D trajectory of a moving car derived from monocular TV-frame sequence of a street scene. Proceedings of IJCAI, pages 692--697, 1992.

....and Rosenfeld used a local quadratic surface fit to find corners. The parameters of the surface were used to find the gradient magnitude and the rate of change of gradient direction; the product of these quantities was used to define cornerness , and local maxima were reported as corners. In [16] Dreschler and Nagel defined corners to be points lying between extrema of Gaussian curvature. Later, for example, see [41] Nagel used the definition (of corner position) of the point of maximum planar curvature for the line of steepest slope . In [76] Zuniga and Haralick found corners at ....

L. Dreschler and H.-H. Nagel. Volumetric Model and 3D Trajectory of a Moving Car Derived from Monocular TV-frame Sequence of a Street Scene. Computer Vision, Graphics and Image Processing, 20(3):199--228, 1981.

....reliably for tens of minutes at a time. One promising approach, using Active Contours [4] uses probabilistic modelling and particle filtering non Gaussian, Monte Carlo inference to deal with visual clutter. An alternative approach extends affine correlation matching [2] and optic flow [7] by using image exemplars [10] for stabilisation. Microphones are in use in any case, for telephony, and it is compelling to try to make use of stereo microphones to help determine the activity and position of speakers, jointly with the analysis of video images. Stereo sound and vision work in ....

L. Dreschler and H.H. Nagel. Volumetric model and 3D trajectory of a moving car derived from monocular TV-frame sequence of a street scene. In Proc. Int. Joint Conf. Artificial Intelligence, pages 692--697, 1981.

....to detect corners via differential geometry. As Noble [49] points out, the determinant of the Hessian matrix and Gaussian CHAPTER 2. REVIEW OF 2D FEATURE DETECTION 13 Equation 1. Following a detailed investigation of the distribution of grey levels about corner features, Dreschler and Nagel [13] also used Beaudet s masks to determine Gaussian curvature , with pixels lying between the extrema of the Gaussian curvature being treated as candidate corner features. Nagel later more clearly defined L junction corner features as the location of the maximum planar curvature in the locus line ....

....The importance of a general model of 2D features is essential to the classification of these features due to the large number of local 2D intensity variations that are 2D features. All previous approaches to 2D feature detection have been shown to model only a subset of these 2D features [2, 38, 3, 13, 75, 28, 11, 60, 5], or to have no model of the features at all [42, 9, 22] In Chapter 4 a general definition of 2D image features is presented that encompasses all types of 2D features, with an implementation of a method for the detection of these features based on this model described in detail in Chapter 5. In ....

L. Dreschler and H.-H. Nagel. Volumetric model and 3-D trajectory of a moving car derived from monocular TV-frame sequence of a street scene. Computer Vision, Graphics, and Image Processing, 20(3):199--228, 1982.

....uniform greyvalues. Therefore, the curvature is multiplied by the gradient magnitude of the image where non maximum suppression is applied to the gradient magnitude before multiplication. Their measure is K = IxxI 2 y Iyy I 2 x Gamma2I xy IxIy I 2 x I 2 y . Dreschler and Nagel [16] first determine locations of local extrema of the determinant of the Hessian DET . A location of maximum positive DET can be matched with a location of extreme negative DET , if the directions of the principal curvatures which have opposite sign 10 are approximatively aligned. The interest ....

L. Dreschler and H.H. Nagel. Volumetric model and 3D trajectory of a moving car derived from monocular TV frame sequences of a street scene. Computer Graphics and Image Processing, 20:199--228, 1982.

....for extrema of K(x, y) normalized by the gradient magnitude over edge pixels only. Beaudet [1] proposed to calculate Hessian determinant H(x, y) f xx f yy f 2 xy (2) of the image function and to find corners as local extrema of this determinant. In Dreschler s and Nagel s approach [3] the local extrema of Gaussian curvature of the image function are identified and corners are localized by interpolation between them. Unlike the above mentioned methods, the corner detector proposed by F orstner [4] uses first order derivatives only. F orstner determines corners as local maxima ....

L. Dreschler and H. Nagel. Volumetric model and 3-D trajectory of a moving car derived from monocular TV-frame sequence of a street scene. Proc. Int. Joint Conf. Artificial Intelligence, pages 692--697, Vancouver, Canada, 1981.

.... [Thorpe, 1984] Marr, Poggio, and Ullman prefer zero crossings of the Laplacian of the image intensity [Marr et al. 1979] and Kitchen, Rosenfeld, Dreschler, and Nagel define corner features based on first and second derivatives of the image intensity function [Kitchen and Rosenfeld, 1980] [Dreschler and Nagel, 1981]. In contrast with these selection criteria, which are defined independently of the registration algorithm, we show in this report that a criterion can be derived that explicitly optimizes the tracking performance. In other words, we define a feature to be good if it can be tracked well. In this ....

L. Dreschler and H.-H. Nagel. Volumetric model and 3d trajectory of a moving car derived from monocular tv frame sequences of a street scene. In Proceedings of the International Joint Conference on Artificial Intelligence, pages 692--697, Vancouver, Canada, August 1981.

....can be classified into a small number of categories depending on the feature they utilize to detect corners. The first category uses the magnitude of the gradient of the gradient direction, at the point of interest as a measure of cornerness. This quantity attains a local maximum at a corner point [7, 8, 12, 13, 30]. This category of approaches has some disadvantages. First, the results are sensitive to noise. Secondly, a priori knowledge of the object size is required. Thirdly, even if the object or the feature size is supplied as an input parameter to the algorithm, the algorithm can not accommodate for ....

L. Drechler and H. Nagel, Volumetric model and 3-d trajectory of a moving car derived from monocular tv-frame sequence of a street scene, in Proceedings IJCAI, pp. 692-697, 1981.

.... [11] possibly with adaptive window size[14] Feature windows can be selected based on some measure of texturedness or cornerness, such as a high standard deviation in the spatial intensity profile [13] the presence of zero crossings of the Laplacian of the image intensity [12] and corners [9] [5]. Yet, even a region rich in texture can be poor. For instance, it can straddle a depth discontinuity or the boundary of a reflection highlight on a glossy surface. In either case, the window is not attached to a fixed point in the world, making that feature useless or even harmful to most ....

L. Dreschler and H.-H. Nagel. Volumetric model and 3d trajectory of a moving car derived from monocular tv frame sequences of a street scene. IJCAI, 692--697, 1981.

....higher resolution estimates. Their advantages include both increased computation efficiency and the ability to find better solutions (escape from local minima) Tracking individual features (corners, points, lines) in images has always been alternative to iconic (pixel based) optic flow techniques [Dreschler and Nagel, 1982; Sethi and Jain, 1987; Zheng and Chellappa, 1992] This has the advantage of requiring less computation and of being less sensitive to lighting variation. The algorithm presented in this paper is closely related to patchbased feature trackers [Lucas and Kanade, 1981; Rehg and Witkin, 1991; Tomasi ....

L. Dreschler and H.-H. Nagel. Volumetric model and 3D trajectory of a moving car derived from monocular tv frame sequences of a stree scene. Computer Graphics and Image Processing, 20:199--228, 1982.

....interest due to their accurate localisation. Sensitivity to noise, speed and selectivity are all issues which need to be considered when attempting to implement a corner detector. A number of methods for detecting corners within shapes include template matching, second order derivative schemes [14, 15], autocorrelation [16] median based methods [17] and the interesting point operator [18] The corner detector implemented in the following experiments is particularly suited to detecting corners which are projections of vertices. It locates corners by first finding regions where several lines are ....

L.Dreschler and H.-H.Nagel. Volumetric model and 3d-trajectory of a moving car derived from monocular tv-frame sequences of a street scene. Proc. Int. Joint Conf. on Artif. Intell., pages 692--697, 1981.

....to find instances of the objects of interest. 3.2 CORNER POINT DETECTION The extraction of distinct points such as corner points has been a basic procedure in digital photogrammetry and computer vision. Previous research in the field of distinct point detection includes work by Moravec (1977) Dreschler and Nagel (1981), Mikhail and Mitchell (1984) Foerstner and Guelch (1987) Foerstner (1994) and Tang and Heipke (1994) The Foerstner operator was chosen for its salient features such as rotational invariance and sub pixel accuracy. A corner point is defined as the intersection of edge elements. The Foerstner ....

Dreschler, L. and Nagel, H. (1981): Volumetric Model and 3D-Trajectory of a Moving Car Derived from Monocular TV-Frame sequences of a Street Scene.

No context found.

Dreschler L., Nagel H.H.; Volumetric model and 3D trajectory of a moving car derived from monocular TV frame sequences of a street scene; Proc. Int. Joint Conf. Artificial Intelligence, 1981, p.692-697.

No context found.

L. Dreschler and H.-H. Nagel. Volumetric Model and 3D Trajectory of a Moving Car Derived from Monocular TV Frame Sequences of a Street Scene. Computer Graphics and Image Processing 20:199-228, 1982.

No context found.

L. Dreschler and H.-H. Nagel. Volumetric Model and 3D Trajectory of a Moving Car Derived from Monocular TV Frame Sequences of a Street Scene. Computer Graphics and Image Processing 20:199-228, 1982.

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