This paper presents the neurocalibration approach as a new neural-based solution for the problem of camera calibration. Unlike some existing neural approaches, our calibrating network can match the perspective-project...
详细信息
This paper presents the neurocalibration approach as a new neural-based solution for the problem of camera calibration. Unlike some existing neural approaches, our calibrating network can match the perspective-projection-transformation matrix between the world 3D points and the corresponding 2D image pixels. Starting from random initial weights, the net can specify the camera model parameters satisfying the orthogonality constraints on the rotational transformation. In order to improve the accuracy of calibration results, the paper demonstrates the application of the neurocalibration technique to multi-image camera calibration. In such a case, many images are taken by the same camera but from different (rotated and/or translated) positions. Experiments have shown the accuracy and the efficiency of our neurocalibration technique.
In this paper, we present a novel and accurate approach for nonrigid registration. New feature descriptors are built as voxel signatures using scale space theory. These descriptors are used to capture the global motio...
详细信息
In this paper, we present a novel and accurate approach for nonrigid registration. New feature descriptors are built as voxel signatures using scale space theory. These descriptors are used to capture the global motion of the imaged object. Local deformations are modelled through an evolution process of equi-spaced closed curves/surfaces (iso-contours/surfaces) which are generated using fast marching level sets and are matched using the built feature descriptors. The performance of the proposed approach is validated using the finite element method. Both 2D and 3D tissue deformations cases are simulated, and the registration accuracy is quantified by co-registering the deformed images with the original ones and comparing the recovered mesh point displacements with the simulated ones. The evaluation results show the potential of the proposed approach in handling local deformation better than some conventional approaches
The structure and properties of several morphological filtering algorithms are discussed. The performance of the algorithms when applied to images corrupted with noise is presented. Several open research problems rela...
详细信息
The structure and properties of several morphological filtering algorithms are discussed. The performance of the algorithms when applied to images corrupted with noise is presented. Several open research problems relative to these nonlinear filters are also discussed.< >
This paper describes design and implementation of a vision based platform for automated refueling tasks. The platform is an autonomous docking system in principle, with the specific application - refueling of vehicles...
详细信息
This paper describes design and implementation of a vision based platform for automated refueling tasks. The platform is an autonomous docking system in principle, with the specific application - refueling of vehicles. The system is. based on monochromatic, monocular vision, and it utilizes very specialized imageprocessing schemes. imageprocessing consists of very fast filtering and segmenting algorithms, as well as moment's computation. A robotic arm with 6 joints (FANUC M-6i), and a controller unit (R-B), does the physical work. A serial interface, with very high-level commands, connects a supercomputing machine and the robot's controller. A practical setup would probably be scaled down to a special design robot, and a single processor, controller with special VLSI chips for imageprocessing. Results are very promising; the robot can identify the cap position, orientation, and height in real time with acceptable accuracy and reliability.
Conventional subspace construction approaches suffer from the need of "large-enough" image ensemble rendering numerical methods intractable. In this paper, we propose an analytic formulation for low-dimensio...
详细信息
ISBN:
(纸本)9781467364102
Conventional subspace construction approaches suffer from the need of "large-enough" image ensemble rendering numerical methods intractable. In this paper, we propose an analytic formulation for low-dimensional subspace construction in which shading cues lie while preserving the natural structure of an image sample. Using the frequency-space representation of the image irradiance equation, the process of finding such subspace is cast as establishing a relation between its principal components and that of a deterministic set of basis functions, termed as irradiance harmonics. Representing images as matrices further lessen the number of parameters to be estimated to define a bilinear projection which maps the image sample to a lower-dimensional bilinear subspace. Results show significant impact on dimensionality reduction with minimal loss of information as well as robustness against noise.
This work deals with non-invasive and non-intrusive measurements of the human facial vasculature from thermal imaging, and estimates the waveforms and rates of the arterial pulse. The paper addresses the issues involv...
详细信息
This work deals with non-invasive and non-intrusive measurements of the human facial vasculature from thermal imaging, and estimates the waveforms and rates of the arterial pulse. The paper addresses the issues involved with use of Long-Wave IR imaging in studies of human biometrics, and focuses on measurements of human's heart rate based on a small number of thermal video frames. Infrared videos of 30 subjects were recorded at two distances. Vascular mapping and multi-resolution analysis are used to calculate a subject's heart rate using only 512 Long-Wave IR video frames (17.07 seconds). This semi-automatic process is typically 89-99% accurate, with average accuracies at both distances at approximately 93%. The work is a step in a comprehensive computervision system for computational studies on human-machine interface.
Stereo computation is one of the vision problems where the presence of outliers cannot be neglected. Most standard algorithms make unrealistic assumptions about noise distributions, which leads to erroneous results th...
详细信息
We consider the problem of computing the 3D shape of an unknown, arbitrary-shaped object. Starting with an initial surface, larger than the scene, we refine it using the object's projections taken at known but arb...
详细信息
We consider the problem of computing the 3D shape of an unknown, arbitrary-shaped object. Starting with an initial surface, larger than the scene, we refine it using the object's projections taken at known but arbitrarily distributed viewpoints. These projections are used independently in a recursive manner. Thus, using more projections improves the reconstruction with a slight increase in the computation time. The technique generates an approximate voxelized representation of the object. We describe the proposed technique and outline the algorithm. Rendered images of voxel spaces recovered from synthetic and real observation images are shown.
暂无评论