Emilio Vital Brazil

Expression Transfer between Photographs through Multilinear AAM's

Expression transfer is a method for mapping a photographed expression performed by a given subject onto the photograph of another persons face. Building on well succeeded previous works by the vision researchers (facial expression decomposition, active appearance models and multilinear analysis, we propose a novel approach for expression transfer based on color images. We attack this problem with methods developed by the computer vision community for facial expression analysis and recognition. Combining active appearance models and multilinear analysis, its possible to suitably represent and analyze expressive facial images, while separating both style (subjects identity) and content (expressive flavor) from the captured performance

iLAMP: Exploring high-dimensional spacing through backward multidimensional projection

Ever improving computing power and technological advances are greatly augmenting data collection and scientific observation. This has directly contributed to increased data complexity and dimensionality, motivating research of exploration techniques for multidimensional data. Consequently, a recent influx of work dedicated to techniques and tools that aid in understanding multidimensional datasets can be observed in many research fields, including biology, engineering, physics and scientific computing. While the effectiveness of existing techniques to analyze the structure and relationships of multidimensional data varies greatly, few techniques provide flexible mechanisms to simultaneously visualize and actively explore high-dimensional spaces. In this paper, we present an inverse linear affine multidimensional projection, coined iLAMP, that enables a novel interactive exploration technique for multidimensional data. iLAMP operates in reverse to traditional projection methods by mapping low-dimensional information into a high-dimensional space. This allows users to extrapolate instances of a multidimensional dataset while exploring a projection of the data to the planar domain. We present experimental results that validate iLAMP, measuring the quality and coherence of the extrapolated data; as well as demonstrate the utility of iLAMP to hypothesize the unexplored regions of a high-dimensional space.

Sketch modeling of seismic horizons from uncertainty

Petroleum reservoir model building is a fundamental but complex task present in all stages of oil/gas exploration and production (E&P). Reservoir models are built incrementally using multi-disciplinary data (e.g. from geophysics, geology, reservoir engineering) and the domain expert interpretation of that data. The first reservoir models are constructed at the appraisal stage, where the available data presents inaccuracies and a high degree of uncertainty. In this paper we present a set of sketch-based interface and modeling operators integrated in a system for the early appraisal stage in oil/gas E&P for the tasks of seismic interpretation and reservoir model building. Our system allows the user to sketch directly over the raw seismic reflection volume and its derived data. These data guide the expert in the key tasks of seismic interpretation and building the structural framework of the reservoir. We propose a novel set of sketch-based modeling operators designed by specific domain requirements from geophysics and geology. A novel architecture using adaptive meshes is also developed to create a more flexible sketch-based system.

Fluid-based hatching for tone mapping in line illustrations

This paper presents a novel meshless, physically-based framework for line art rendering of surfaces with complex geometry and arbitrary topology. We apply an inviscid fluid flow simulation using Smoothed Particles Hydrodynamics to compute the global velocity and cross fields over the surface model. These fields guide the automatic placement of strokes while extracting the geometric and topological coherence of the model. Target tones are matched by tonal value maps allowing different hatching and cross-hatching effects. We demonstrate the simplicity and effectiveness of our method with sample renderings obtained for a variety of models.

Extended papers from NPAR 2010: Shape and tone depiction for implicit surfaces

We present techniques for rendering implicit surfaces in different pen-and-ink styles. The implicit models are rendered using point-based primitives to depict shape and tone using silhouettes with hidden-line attenuation, drawing directions, and stippling. We present sample renderings obtained for a variety of models. Furthermore, we describe simple and novel methods to control point placement and rendering style. Our approach is implemented using HRBF Implicits, a simple and compact representation, that has three fundamental qualities: a small number of point-normal samples as input for surface reconstruction, good projection of points near the surface, and smoothness of the gradient field. These qualities of HRBF Implicits are used to generate a robust distribution of points to position the drawing primitives.

Sketch-based warping of RGBN images

While current image deformation methods are careful in making the new geometry seem right, little attention has been given to the photometric aspects. We introduce a deformation method that results in coherently illuminated objects. For this task, we use RGBN images to support a relighting step integrated in a sketch-based deformation method. We warp not only colors but also normals. Normal warping requires smooth warping fields. We use sketches to specify sparse warping samples and impose additional constraints for region of interest control. To satisfy these new constraints, we present a novel image warping method based on Hermite-Birkhoff interpolation with radial basis functions that results in a smooth warping field. We also use sketches to help the system identify both lighting conditions and material from single images. We present results with RGBN images from different sources, including photometric stereo, synthetic images, and photographs.

Interactive cutaways of oil reservoirs

Abstract In the Oil and Gas industry, processing and visualizing 3D models is of paramount importance for making exploratory and production decisions. Hydrocarbons reservoirs are entities buried deep in the earth’s crust, and a simplified 3D geological model that mimics this environment is generated to run simulations and help understand geological and physical concepts. For the task of visually inspecting these models, we advocate the use of Cutaways: an illustrative technique to emphasize important structures or parts of the model by selectively discarding occluding parts, while keeping the contextual information. However, the complexity of reservoir models imposes severe restrictions and limitations when using generic illustrative techniques previously proposed by the computer graphics community. To overcome this challenge, we propose an interactive Cutaway method, strongly relying on screen-space GPU techniques, specially designed for inspecting 3D reservoir models represented as corner-point grids, the industry’s standard.

Facing the high-dimensions

Multidimensional projection has become a standard tool for visual analysis of multidimensional data sets, as the 2D representation of multidimensional instances gives an important and informative panorama of the data. Recently, research in this topic has been steered towards methods that permit user intervention and interactivity. One of such methods is inverse projection, a recently proposed resampling mechanism that allows users to generate new multidimensional instances by creating reference 2D points in the projection space. Given an m-dimensional data set and its 2D projection, inverse projection transforms a user-defined 2D point into an m-dimensional point by means of a mapping function. In this work, we propose a novel inverse projection technique based on Radial Basis Functions interpolation. Our technique provides a smooth and global mapping from low to high dimensions, in contrast with the former technique (iLAMP) which is local and piecewise continuous. In order to demonstrate the potential of our technique, we use a 3D human-faces data set and a procedure to interactively reconstruct and generate new 3D faces. The results demonstrate the simplicity, robustness and efficiency of our approach to create new face models from a structured data set, a task that would typically require the manipulation of hundreds of parameters. Graphical abstractDisplay Omitted HighlightsWe develop a new inverse projection method with Radial Basis Functions interpolation.A face-synthesis application using the proposed method to create new face models.We present a workflow to create new expressions and face models from existing ones.An interface to create faces based on the visual feedback of a 2D projection space.Expression transfer mechanism to create the various expressions of a new character.

Interacting with microseismic visualizations

Microseismic visualization systems present complex 3D data of small seismic events within oil reservoirs to allow experts to explore and interact with that data. Yet existing systems suffer several problems: 3D spatial navigation and orientation is difficult, and selecting 3D data is challenging due to the problems of occlusion and lack of depth perception. Our work mitigates these problems by applying both proxemic interactions and a spatial input device to simplify how experts navigate through the visualization, and a painting metaphor to simplify how they select that information.

A few good samples: shape & tone depiction for Hermite RBF implicits

We present techniques for rendering Hermite Radial Basis Function (HRBF) Implicits in different pen-and-ink styles. HRBF Implicits is a simple and compact representation, providing three fundamental qualities: a small number of point-normal samples as input for surface reconstruction, good projection of points near the surface, and smoothness of the gradient field. Our approach uses these qualities of HRBF implicits to generate a robust distribution of points to position the drawing primitives. The resulting implicit model is then rendered using point-based primitives to depict shape and tone using silhouettes with hidden-line attenuation, drawing directions, and stippling. We present sample renderings obtained for a variety of models.