A region-based reconstruction technique is proposed which simplifies the process of scene reconstruction and increases reliability. Integrating grid coding and intensity image analysis methods, stable regions which ma...
详细信息
ISBN:
(纸本)0818620579
A region-based reconstruction technique is proposed which simplifies the process of scene reconstruction and increases reliability. Integrating grid coding and intensity image analysis methods, stable regions which make up the skeleton structure of a room (i.e. wall, ceiling, floor, etc.) are extracted. The indoor scene is then reconstructed based on 2-D and 3-D topological relationships. The experimental results show that local errors do not radically alter the topological relationships of detected regions, so robust reconstruction can be achieved.
A general method is presented for recovering straight homogeneous generalized cylinders from monocular intensity images. In this method, it is assumed that the generalized cylinder being recovered has purely diffuse r...
详细信息
ISBN:
(纸本)0818620617
A general method is presented for recovering straight homogeneous generalized cylinders from monocular intensity images. In this method, it is assumed that the generalized cylinder being recovered has purely diffuse reflectance and that the diffuse reflectance coefficient is constant. It is demonstrated that contour information alone is insufficient to recover a straight homogeneous generalized cylinder uniquely. It is shown that the sign and magnitude of the Gaussian curvature at a point vary among members of a contour-equivalent class. The contour image fails to constrain two parameters of the underlying generalized cylinder, the 3-D axis tilt and translation. A method for ruling straight homogeneous generalized cylinder images is described. Once the rulings of the image have been recovered, all parameters derivable from contour alone can be recovered. To recover the two remaining parameters (modulo scale) not constrained by image contour, additional information must be incorporated into the recovery process, e.g., intensity information. A method for recovering the tilt of the object using the ruled contour image and intensity values along extremal cross-section curves is derived, along with a method for recovering the location of the object's 3-D axis from intensity values along meridians of the surface. The different methods outlined constitute an algorithm for recovering all the shape parameters (modulo scale) of the straight homogeneous generalized cylinder.
A representation scheme is presented for shapes of three-dimensional curved objects whose surface curvatures are not continuous at all points, i.e., surfaces with C1 continuity only. A class of surface curves and poin...
详细信息
ISBN:
(纸本)0818620579
A representation scheme is presented for shapes of three-dimensional curved objects whose surface curvatures are not continuous at all points, i.e., surfaces with C1 continuity only. A class of surface curves and points called surface structure descriptors for this representation is defined. These descriptors consist of subsets of lines of curvature and their intersections and are based on local descriptions of smooth surfaces. The surface structure curves are derived from the zero-crossings, asymptotes and gradient directions of the Gaussian and mean curvatures of the surface. A smooth surface sketch in terms of the geometrical structures of enclosing boundaries of bumps and dents, ridge and valley lines, and peaks and pits (curvature extremum points) is developed. A hierarchical description of curved surfaces that leads to a global shape representation of the object is obtained from these descriptors. To handle the discontinuities of surface curvatures, a computational scheme is introduced in which gradient directions of surface curvatures at the discontinuities can be inferred. Simulation results on the computation of these descriptors from curved surfaces are shown. Potential applications of these representation schemes in computer vision tasks are discussed.
A special problem in nonrigid motion analysis, the problem of estimating the motion and deformation of the human heart, is addressed. The research has been conducted on the left ventricle 3-D data of the bifurcation p...
详细信息
ISBN:
(纸本)0818620579
A special problem in nonrigid motion analysis, the problem of estimating the motion and deformation of the human heart, is addressed. The research has been conducted on the left ventricle 3-D data of the bifurcation points, obtained within the period of one cardiac cycle. The process of the motion analysis is divided into three phases: global motion estimation, global motion compensation, and local deformation estimation. The global motion of the left ventricle is determined using the estimated principal axes over successive time instants. Upon compensation for the global motion, the local deformation is analyzed utilizing the correspondence of individual bifurcation points over the period of the cardiac cycle via tensor analysis. Finally, in order to better understand the complex nature of the left ventricle movement, several scientific visualization techniques are exploited to view the dynamic deformation of the left ventricle surface.
The authors present a model of deformation which solves some of the problems encountered with the original method such as instability and initial data while reducing the computational complexity. This model makes the ...
详细信息
ISBN:
(纸本)0818620579
The authors present a model of deformation which solves some of the problems encountered with the original method such as instability and initial data while reducing the computational complexity. This model makes the curve behave like a balloon which is inflated by an additional force. The initial curve need no longer be close to the solution to converge. The external forces that push the curve to the edges are modified to give more stable results. The system is solved using a conform finite element method in the minimization process. The evolution to the equilibrium presents less oscillations, convergence is obtained faster, and the final results are more accurate. This model is applied for segmenting ultrasound and magnetic resonance images. The authors have also made a first stage to 3-D object reconstruction, by tracking the extracted contour on a series of successive cross sections.
A standard probabilistic formulation of a local shape estimator due to Witkin (1980) has been demonstrated to work to some extent on a range of data, but it also has been shown to overestimate slant. Here, the author ...
详细信息
ISBN:
(纸本)081861952x
A standard probabilistic formulation of a local shape estimator due to Witkin (1980) has been demonstrated to work to some extent on a range of data, but it also has been shown to overestimate slant. Here, the author identifies and analyzes three sources of this error: the choice of prior probability distribution on shape;as systematic bias in the observation of image events;and a kind of nonergodicity in the stochastic process associated with surface events. As a result of this analysis, the author formulates a corrected probabilistic local shape estimator, proves a correctness theorem for it, and demonstrates its superiority to the Wilkin estimator in a set of experiments.
Using a model of an edge's motion through a sequence of images, the problem of its localization can be formulated as a stochastic filtering problem. The extended Kalman filter for such a system is considered in de...
详细信息
ISBN:
(纸本)081861952x
Using a model of an edge's motion through a sequence of images, the problem of its localization can be formulated as a stochastic filtering problem. The extended Kalman filter for such a system is considered in detail and is shown to be interpretable as a sequence of oriented special convolutions. Results are presented which show that the edge localization obtained using this filter is substantially better than that obtained using the either the Sobel or Canny edge operators on each image individually.
The proceedings contains 97 papers. The following topics are dealt with: edge detection;shape derivation;feature extraction;motion;morphology and neural networks;range data;image and texture segmentation;polarization ...
详细信息
ISBN:
(纸本)081861952x
The proceedings contains 97 papers. The following topics are dealt with: edge detection;shape derivation;feature extraction;motion;morphology and neural networks;range data;image and texture segmentation;polarization cues;stereo vision;methodology and standards;object recognition;navigation;preprocessing and computer architectures.
A method to reduce image resolution that simulates the reduction of shape resolution is presented. Such resolution reduction is critical for shape-from-shading algorithms. This method of resolution reduction is used i...
详细信息
ISBN:
(纸本)081861952x
A method to reduce image resolution that simulates the reduction of shape resolution is presented. Such resolution reduction is critical for shape-from-shading algorithms. This method of resolution reduction is used in accelerating the shape from shading by using small multiresolution pyramids. A Lambertian reflectance model is used, but a similar approach can be developed for other reflectance models. The goal is to show that the resolution reduction based only on image intensities is inferior to resolution reduction using knowledge on the surface reflectance.
The authors report theoretical and experimental results which underline the importance of mutual illumination to visual modules dealing with shape and with surface lightness. The experiments are in good agreement with...
详细信息
ISBN:
(纸本)081861952x
The authors report theoretical and experimental results which underline the importance of mutual illumination to visual modules dealing with shape and with surface lightness. The experiments are in good agreement with results obtained with a simple theoretical model. These results show the effects of mutual illumination in pictures of simple objects, and indicate that these effects must be accounted for in modeling image intensities. The data imply that shape from shading based on the image irradiance equation make real errors on images of concave objects, and that edge detectors that respond to only step edges perform badly on polyhedral scenes and waste information.
暂无评论