In recent years, many systems have been developed to handle real-time rendering of a 3D-city model and its terrain. These terrains are often constructed using one of the two main methods for representing terrain e.g. ...
详细信息
A method to detect boundaries in in natural color images is here proposed, combining edge information and region information. This unsupervised fully automatic process uses edge map information to eliminate false boun...
详细信息
A method to detect boundaries in in natural color images is here proposed, combining edge information and region information. This unsupervised fully automatic process uses edge map information to eliminate false boundaries in the image region map, and region map information to remove noise in the image edge map. Thus, it integrates these two maps into a single one to get the final result. This proposal is extensively compared to the multi-label graph cut approach, since both approaches are unsupervised and fully automatic, as well as receive the same two inputs, although performing different processing. Experiments performed on a large set of natural color images were the base for such comparison. The results show that the approach here proposed is promising, besides allowing interesting interpretations about boundary detection.
The surface splatting technique has several advantages in rendering models consisting of points, such as speed and quality of rendered images. After projecting all the splats on the plane of the image, it is necessary...
详细信息
Object detection is a computer Vision task that determines if there is an object of some category (class) in an image or video sequence. When the classes are formed by only one specific object, person or place, the ta...
详细信息
Object detection is a computer Vision task that determines if there is an object of some category (class) in an image or video sequence. When the classes are formed by only one specific object, person or place, the task is known as instance detection. Object recognition classifies an object as belonging to a class in a set of known classes. In this work we deal with an instance detection/recognition task. We collected pictures of famous landmarks from the Internet to build the instance classes and test our framework. Some examples of the classes are: monuments, churches, ancient constructions or modern buildings. We tested several approaches to the problem and a new global feature is proposed to be combined to some widely known features like PHOW. A combination of features and classifiers to model the given instances in the training phase was the most successful one.
Scientific visualization techniques create images attempting to reveal complex structures and phenomena. Illustrative techniques have been incorporated to scientific visualization systems in order to improve the expre...
详细信息
Scientific visualization techniques create images attempting to reveal complex structures and phenomena. Illustrative techniques have been incorporated to scientific visualization systems in order to improve the expressiveness of such images. The rendering of feature lines is an important technique for better conveying surface shapes. In this paper, we propose to combine volume visualization of unstructured meshes with direct rendering of illustrated iso surfaces. This is accomplished by extending a GPU-based ray-casting algorithm to incorporate illustration with photic extremum lines, a type of feature lines that captures sudden change of luminance, conveying shapes in a perceptually correct way.
Important engineering applications use unstructured hexahedral meshes for numerical simulations. Hexahedral cells, when compared to tetrahedral ones, tend to be more numerically stable and to require less mesh refinem...
详细信息
Important engineering applications use unstructured hexahedral meshes for numerical simulations. Hexahedral cells, when compared to tetrahedral ones, tend to be more numerically stable and to require less mesh refinement. However, volume visualization of unstructured hexahedral meshes is challenging due to the trilinear variation of scalar fields inside the cells. The conventional solution consists in subdividing each hexahedral cell into five or six tetrahedra, approximating a trilinear variation by a piecewise linear function. This results in inaccurate images and increases the memory consumption. In this paper, we present an accurate ray-casting volume rendering algorithm for unstructured hexahedral meshes. In order to capture the trilinear variation along the ray, we propose the use of quadrature integration. A set of computational experiments demonstrates that our proposal produces accurate results, with reduced memory footprint. The entire algorithm is implemented on graphics cards, ensuring competitive performance.
Seamless image cloning has become one of the most important editing operation for photomontage. Recent coordinate-based methods have lessened considerably the computational cost of image cloning, thus enabling interac...
详细信息
Seamless image cloning has become one of the most important editing operation for photomontage. Recent coordinate-based methods have lessened considerably the computational cost of image cloning, thus enabling interactive applications. However, those techniques still bear severe limitations as to concavities and dynamic shape deformation. In this paper we present novel methodology for image cloning that turns out to be highly efficient in terms of computational times while still being more flexible than existing techniques. Our approach builds on combinatorial Laplacian and fast Cholesky factorization to ensure interactive image manipulation, handling holes, concavities, and dynamic deformations during the cloning process. The provided experimental results show that the proposed technique outperforms existing methods in requisites such as accuracy and flexibility.
Content-based image classification/retrieval based on image descriptors has become an essential component in most database systems. However, most existing systems do not provide mechanisms that enable interactive mult...
详细信息
Content-based image classification/retrieval based on image descriptors has become an essential component in most database systems. However, most existing systems do not provide mechanisms that enable interactive multi-objective queries, hampering the user experience. In this paper we present a novel methodology capable of accomplishing multi-objective searches while still being interactive. Our approach relies on a combination of class-specific metrics and multidimensional projection to devise an effective and interactive image retrieval system. Besides allowing visual exploration of image data sets, the provided results and comparisons show that the proposed approach outperforms existing methods, turning out to be a very attractive alternative for managing image data sets.
The resolution enhancement of an image is always desirable, for almost all situations, but mainly if the image has the purpose of visual analysis. The hardware development for increasing the image resolution at its ca...
详细信息
The resolution enhancement of an image is always desirable, for almost all situations, but mainly if the image has the purpose of visual analysis. The hardware development for increasing the image resolution at its capture still has a higher cost than the algorithm solutions for super-resolution (SR). Like image restoration, super-resolution is also an ill-conditioned inverse problem. This work analyses the iterative restoration methods (Van Cittert, Tikhonov-Miller and Conjugate Gradient) which propose solutions for the ill-conditioning problem and compares them with the IBP method (Iterative Back Projection) proposed by Irani-Peleg [1] and Komatsu et al. [2]. The analysis of the found similarities is the basis of a generalization, that other iterative restoration methods can have their properties adapted, such as regularization of the ill-conditioning, noise reduction and other degradations and the increase of the convergence rate can be incorporated to the techniques of super-resolution.
In recent years, GPUs (graphicsprocessing Units) have shown a significant advance of computational resources available for the use of non-graphical applications. The ability to solve problems involving parallel compu...
详细信息
In recent years, GPUs (graphicsprocessing Units) have shown a significant advance of computational resources available for the use of non-graphical applications. The ability to solve problems involving parallel computing as well as the development of new architectures that supports this new paradigm, such as CUDA (Compute Unified Device Architecture), have encouraged the use of GPU for general purpose applications, especially in games. Some parallel tasks which were CPU (Central processing Unit) based are being ported over to the GPU due to theirs superior performance. One of these tasks is the path finding of an agent over a game map, which has already achieved a better performance on GPU, but is still limited. This paper describes some optimizations to a GPU path finding implementation, addressing larger work set (agents and nodes) with good performance.
暂无评论