terça-feira, 8 de janeiro de 2013

2009f - Scale-Space for Union of 3D Balls

Abstract—Shape discretization through union of weighted
points or balls appears as a common representation in different
fields of computer graphics and geometric modeling.
Among others, it has been very successful for implicit surface
reconstruction with radial basis functions, molecular atomic
models, fluid simulation from particle systems and deformation
tracking with particle filters. These representations are
commonly generated from real measurements or numerical
computations, which may require filtering and smoothing operations.
This work proposes a smoothing mechanism for union
of balls that tries to inherit from the scale-space properties
of bi-dimensional curvature motion: it avoids disconnecting
the shape, prevents self-intersection, regularly decreases the
area and convexifies the shape. The smoothing is computed
iteratively by moving each ball of the union according to a
combination of projected planar curvature motions. Experiments
exhibits nice properties of this scale-space.
Keywords-Union of Balls; Scale Spaces; Curvature Motion;

2009e - Geometry Super-Resolution by Example

Abstract—The acquisition of high-resolution 3D models still
requires delicate and time-consuming processes. In particular,
each detail of the object should be scanned separately, although
they may be similar. This can be simplified by copying a small
set of details at different places of the model, synthesizing
high geometric resolution from details exemplars, as introduced
in this paper for three different contexts : when the detail
exemplars are scanned separately at high resolution, when
they are synthesized or edited from other models, or when
they are obtained by accumulating repeated instances of the
detail in the low-resolution scan. The main challenge here is
to correctly register the high-resolution details with the low
resolution model. To address this issue, this work proposes
a careful resolution manipulation of 3D scans at each step
of an automatic registration pipeline, combined with a robust
selection of alignments. This results in a fully automatic process
for geometry super-resolution by example. Experiments on synthetic
and real data sets show applicability in different contexts,
including resolution increase, noise removal by example and
geometric texture insertion.

2009d - Support Vectors Learning for Vector Field Reconstruction Marcos

Abstract—Sampled vector fields generally appear as measurements
of real phenomena. They can be obtained by the
use of a Particle Image Velocimetry acquisition device, or
as the result of a physical simulation, such as a fluid flow
simulation, among many examples. This paper proposes to
formulate the unstructured vector field reconstruction and
approximation through Machine-Learning. The machine learns
from the samples a global vector field estimation function
that could be evaluated at arbitrary points from the whole
domain. Using an adaptation of the Support Vector Regression
method for multi-scale analysis, the proposed method provides
a global, analytical expression for the reconstructed vector field
through an efficient non-linear optimization. Experiments on
artificial and real data show a statistically robust behavior of
the proposed technique.
Keywords-Discrete Vector Field; Support Vector Machine;

2009c - Random Walks for Vector Field Denoising

Abstract—In recent years, several devices allow to directly
measure real vector fields, leading to a better understanding
of fundamental phenomena such as fluid simulation or brain
water movement. This turns vector field visualization and
analysis important tools for many applications in engineering
and in medicine. However, real data is generally corrupted
by noise, puzzling the understanding provided by those tools.
Those tools thus need a denoising step as preprocessing,
although usual denoising removes discontinuities, which are
fundamental for vector field analysis. This paper proposes a
novel method for vector field denoising based on random walks
which preserve those discontinuities. It works in a meshless
setting; it is fast, simple to implement, and shows a better
performance than the traditional gaussian denoising technique.
Keywords-Discrete Vector Field; Denoising; Random Walk;
Markov Chain;

2009b - Learning good views through intelligent galleries

Abstract. The definition of a good view of a 3D scene is highly subjective and strongly depends on both the scene
content and the 3D application. Usually, camera placement is performed directly by the user, and that task may
be laborious. Existing automatic virtual cameras guide the user by optimizing a single rule, e.g. maximizing the
visible silhouette or the projected area. However, the use of a static pre-defined rule may fail in respecting the
user’s subjective understanding of the scene. This work introduces intelligent design galleries, a learning approach
for subjective problems such as the camera placement. The interaction of the user with a design gallery teaches
a statistical learning machine. The trained machine can then imitate the user, either by pre-selecting good views
or by automatically placing the camera. The learning process relies on a Support Vector Machines for classifying
views from a collection of descriptors, ranging from 2D image quality to 3D features visibility. Experiments of the
automatic camera placement demonstrate that the proposed technique is efficient and handles scenes with occlusion
and high depth complexities. This work also includes user validations of the intelligent gallery interface.
Keywords: Learning. Camera Positioning. Virtual Camera. Intelligent Camera. Good View.

2009a - Arch generated shear bands in granular systems

Abstract. We propose an arch based model, on cubic and square lattices, to simulate the internal mobility of
grains, in a dense granular system under shear. In this model, the role of the arches in granular transport presents
a non-linear dependence on the local values of the stress components that can be modeled geometrically. This
non-linearity is very important since a linear dependence on the stress will make the models behave similarly to
viscous fluids, which will not reproduce highly interesting properties of the sheared systems such as shear bands. In
special, we study a modified Couette flow and find the appearance of shear-bands in accordance with the literature.
Keywords: Shear band. Granular system. Transport properties. Bi-based cuprates.

2008b - Approximations by smooth transitions in binary space partitions

result without global optimization. It combines the flexibility of Binary Space Partitions Trees with the statistical
robustness of Smooth Transition Regression Trees. The construction of the tree is straightforward and easily
controllable, using error-driven metrics or external constraints. Moreover, it leads to a concise representation.
Applications on synthetic and real data, both scalar and vector-valued demonstrated the effectiveness of this
Keywords: Binary Space Partitions. Smooth Transition Regression Trees. Classification and Regression Tree.
Vector Fields Approximation. Curve Reconstruction. Geometric Modeling.

2008a - Statistical optimization of octree searches

This work emerged from the following observation: usual search procedures for octrees start from the
root to retrieve the data stored at the leaves. But since the leaves are the farthest nodes to the root, why start from the root? With usual octree representations, there is no other way to access a leaf. However, hashed octrees allow direct access to any node, given its position in space and its depth in the octree. Search procedures take the position as an input, but the depth remains unknown. This work proposes to estimate the depth of an arbitrary node through a statistical optimization of the average cost of search procedures. Since the highest costs of these algorithms are obtained when starting from the root, this method improves on both the memory footprint by the use of hashed octrees, and execution time through the proposed optimization.
Keywords: Octree. Hashing. Quadtree. Geometric Modelling. Data Structures.

2007f - Mineração Visual de Multiatributos Sísmicos para Classificação de Multifácies

The understanding of large data clustering is a relevant problem in several areas of science and engineering. Seismic signal are in general represented by large data sets. For petroleum geophysicists the choice of areas of interest is a demanding task. In this paper we introduce data mining visualizationtechniques, stellar coordinates and transfer functions, which easy the task of dealing visually with data clustering in an interactive and intuitive way allowing the user to make meaningful choices of 3D regions of interest to work with.

2007e - Reconstrução Topológica Tridimensional de Objetos Geológicos em Bacias Marítimas Usando Dados Sísmicos 2D ou 3D: uma Ferramenta Exploratória e Explotatória de Hidrocarbonetos

The interplay between geophysics and geology is today one of the main challenges in reservoir characterization. In this paper we expose the ideia that by careful mathematical formulation and computer graphics tools it is possible to reconstruct 3D geological forms starting with seismic parameters. We reconstruct all the topological components of isosurfaces in subvoxel resolution starting with seismic parameters. Each isosurface component represents the boundary of a geobody. This type of reconstruction helps the modeler to understanding better the seismic information, the geobody spatial characteristics and make quantitative estimations such as reservoir volume.

2007d - Métodos Baseados em Núcleos e Máquinas de Suporte Vetorial em Aplicações de Geofísica de Petróleo

Kernel based methods and support vector machines are recognized today as important methods in areas such as artificial intelligence and bioinformatics, among others. Increasingly in the last years applications to geophysics is becoming widespread, emphasis made to the oil industry. In this paper we propose a new method to distinguish between reservoir and non reservoir characteristics in a seismic sample. Volumetric multiatribute seismic and well data are used in the training process.

2007c - Data Inspection on Multiatribute Seismic Modeling using Star Coordinates

One purpose of multi-attribute data visualization is to display a data set in a manner where
the user can easily ¯nd some kind of underlay structure. The main challenge resides in the fact
that there are only two dimensions to project in a display. In this work we are proposing the
use of star coordinates as a tool for modeling and visualization of multi-attributes seismic data.
Such tool shows to be very useful since it allows the user not only to identify visually clusters
but also correlations between the attributes. The main objective of this work is to develop a
simple interactive visual tool that facilitates multivariate seismic data exploration and analysis.
The software is a Gocad plugin.

2007b - Multi-attribute Seismic Cell Reservoir/Non-Reservoir Classi¯cation with Kernel Based Methods

Kernel machines are recognized as an important class of arti¯cial intelligence methods. Their
use in geosciences applications have been increasing in recent years. In this paper we propose a new
scheme to distinguish each seismic cell as reservoir or non-reservoir. For this, we use multi-attribute
seismic and well data for the learning process. The method shows to be very promising and by the
use of a friendly GoCAD interface the user specify the few parameters of the method.

2007a - Reservoir reconstruction from seismic data with topological guarantees

This paper describes the main aspects of the Geosis-BR GoCAD plugin. It is an ongoing three
year project between the Brazilian oil company Petrobras and Matmidia Laboratory Pontifical
Catholic University of Rio de Janeiro, Brazil. The conceptual core of the software is to extract reli-
able reservoir information from seismic data through the use of geometric and topological modeling,
as well as scientific visualization.

2006 - Article - Exploratory visualization based on multidimensional transfer functions and star coordinates

Abstract: Exploration and analysis of multivariate data play an important role in different domains. This work proposes a simple interface prototype that allows a human beimg to explore visually multivariate spatial objects, like image, sequence of images or volume. It uses star coordinates to display the multivariate data on the computer 2D screen. Once a user identifies a feature on this powerful coordinate system, he/she maps a selected feature region on such widget to a color and an opacity. As a visual result, the feature is rendered on the object space domain by the use of these maps. We also show some examples that illustrates the interface potencial to some applications.

2006 - Article - Point set compression through BSP quantization

Abstract: This work introduces a new compression scheme for point sets. This scheme relies on an adaptive binary space partition (BSP) which takes into account the geometric structure of the point set. This choice introduces geometrical rather than combinatorial information in the compression scheme. In order to effectively improve the final compression ratio, this partition is encoded in a progressive manner, decreasing the number of bits used for the quantisation at each subdivision. This strategy distributes the extra cost of the geometry encoding onto the maximal number of points, compressing in average 15% more than previous techniques.

Keywords: Point Sets, Compression, Binary Space Partition, Geometry-Driven Compression, Geometry Processing.

2006 - Tecnical Report - Real Time Fluid Dynamics on Programmable GPU with Arbitrary Boundaries and Vorticity Confinement

Abstract: In this work, we will present a method to solve the incompressible Navier-Stokes equations using graphics hardware to simulate the fluid behavior. The method is fast enough to allow interaction with the fluid even on large domains. Using a new approach in the treatment of the boundary we get good results in simulations over complex domains. The fluid simulation is based on the method described in [18],and to solve the problem with numerical dissipation we present a new scheme in GPU for vorticity confinement. The implementation is made with Cg.

2006 - Master Thesis: Navier Stokes in GPU - Alex Laier Bordignon

Resumo: Nesse trabalho, mostramos como simular um fluido em duas dimensões em um domínio com fronteiras arbitrárias. Nosso trabalho é baseado no esquema stable fluids desenvolvido por Joe Stam. A implementação é feita na GPU (Graphics Processing Unit), permitindo velocidade de interação com o fluido. Fazemos uso da linguagem Gg (C for Graphics), desenvolvida pela companhia NVidia. Nossas principais contribuições são o tratamento das múltiplas fronteiras, onde aplicamos interpolação bilinear para atingir melhores resultados, armazenamento das condições de fronteira usa apenas um canal de textura, e o uso de confinamento de vorticidade.