Journal of the Brazilian Computer Society, Apr 1, 1997
We discuss the problem of adaptive polygonization of regular surfaces of the euclidean 3D space, ... more We discuss the problem of adaptive polygonization of regular surfaces of the euclidean 3D space, and present e ective algorithms for computing optimal polygonizations of surfaces described in parametric or implicit form.
Figure 1: The collecTable running on the iTable (a), with several fiducials over it. The digital ... more Figure 1: The collecTable running on the iTable (a), with several fiducials over it. The digital projections of the fiducials floating over the interface (b), each fiducial storing different music collections. The M-Cube interface for music data showing albums (c) and tracks (d).
The novelty of our proposal is the end-to-end solution to combine computer generated elements and... more The novelty of our proposal is the end-to-end solution to combine computer generated elements and captured panoramas. This framework supports productions specially aimed at spherical displays (e.g., fulldomes). Full panoramas are popular in the computer graphics industry. However their common usage on environment lighting and reflection maps are often restrict to conventional displays. With a keen eye in what may be the next trend in the filmmaking industry, we address the particularities of those productions, exploring a new representation of the space by storing the depth together with the light map, in a full panoramic light-depth map.
We show how to use affine arithmetic to represent a parametric curve with a strip tree. The requi... more We show how to use affine arithmetic to represent a parametric curve with a strip tree. The required bounding rectangles for pieces of the curve are computed by exploiting the linear correlation information given by affine arithmetic. As an application, we show how to compute approximate distance fields for parametric curves.
In this paper, we propose the use of traditional animations, heuristic behavior and reinforcement... more In this paper, we propose the use of traditional animations, heuristic behavior and reinforcement learning in the creation of intelligent characters for computational media. The traditional animation and heuristic gives artistic control over the behavior while the reinforcement learning adds generalization. The use case presented is a dog character with a high-level controller in a 3D environment which is built around the desired behaviors to be learned, such as fetching an item. As the development of the environment is the key for learning, further analysis is conducted of how to build those learning environments, the effects of environment and agent modeling choices, training procedures and generalization of the learned behavior. This analysis builds insight of the aforementioned factors and may serve as guide in the development of environments in general.
Constructing high-quality generative models for 3D shapes is a fundamental task in computer visio... more Constructing high-quality generative models for 3D shapes is a fundamental task in computer vision with diverse applications in geometry processing, engineering, and design. Despite the recent progress in deep generative modelling, synthesis of finely detailed 3D surfaces, such as high-resolution point clouds, from scratch has not been achieved with existing approaches. In this work, we propose to employ the latent-space Laplacian pyramid representation within a hierarchical generative model for 3D point clouds. We combine the recently proposed latent-space GAN and Laplacian GAN architectures to form a multi-scale model capable of generating 3D point clouds at increasing levels of detail. Our evaluation demonstrates that our model outperforms the existing generative models for 3D point clouds.
Figure 1. Overview of our method: from a 2D photo and their corresponding facial landmarks (a)-(b... more Figure 1. Overview of our method: from a 2D photo and their corresponding facial landmarks (a)-(b), the facial texture data is extracted by successive 2D triangular subdivisions (c)-(d), producing a new 3D face model (e)-(f).
We introduce a novel framework for automatic 3D facial expression analysis in videos. Preliminary... more We introduce a novel framework for automatic 3D facial expression analysis in videos. Preliminary results demonstrate editing facial expression with facial expression recognition. We first build a 3D expression database to learn the expression space of a human face. The real-time 3D video data were captured by a camera/projector scanning system. From this database, we extract the geometry deformation independent of pose and illumination changes. All possible facial deformations of an individual make a nonlinear manifold embedded in a high dimensional space. To combine the manifolds of different subjects that vary significantly and are usually hard to align, we transfer the facial deformations in all training videos to one standard model. Lipschitz embedding embeds the normalized deformation of the standard model in a low dimensional generalized manifold. We learn a probabilistic expression model on the generalized manifold. To edit a facial expression of a new subject in 3D videos, the system searches over this generalized manifold for optimal replacement with the 'target' expression, which will be blended with the deformation in the previous frames to synthesize images of the new expression with the current head pose. Experimental results show that our method works effectively.
Region-based approaches to cel painting typically use shape similarity and topology relations bet... more Region-based approaches to cel painting typically use shape similarity and topology relations between regions of consecutive animation frames. This paper presents a new colorization algorithm based on topological differences defined over a hierarchical graph of adjacent regions, which allows an almost full automatic colorization process. Also this paper discusses other attributes that improve the solution of the image association problem.
In this paper we introduce an unified framework for basic operations on combinatorial 2manifolds ... more In this paper we introduce an unified framework for basic operations on combinatorial 2manifolds with or without boundary. We show that there are two kinds of primitive operators on the underlying meshes: operators which change the topological characteristic of the mesh and operators which just modify its combinatorial structure. We present such operators and demonstrate that they provide a complete set of elementary operations for mesh modification. We also give a description of the algorithms and data structures for an efficient implementation of these operators.
2022 35th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)
We present MR-Net, a general architecture for multiresolution neural networks, and a framework fo... more We present MR-Net, a general architecture for multiresolution neural networks, and a framework for imaging applications based on this architecture. Our coordinate-based networks are continuous both in space and in scale as they are composed of multiple stages that progressively add finer details. Besides that, they are a compact and efficient representation. We show examples of multiresolution image representation and applications to texture magnification, minification, and antialiasing. This document is the extended version of the paper [PNS + 22]. It includes additional material that would not fit the page limitations of the conference track for publication.
We introduce a neural implicit framework that exploits the differentiable properties of neural ne... more We introduce a neural implicit framework that exploits the differentiable properties of neural networks and the discrete geometry of point-sampled surfaces to approximate them as the level sets of neural implicit functions. To train a neural implicit function, we propose a loss functional that approximates a signed distance function, and allows terms with high-order derivatives, such as the alignment between the principal directions of curvature, to learn more geometric details. During training, we consider a non-uniform sampling strategy based on the curvatures of the point-sampled surface to prioritize points with more geometric details. This sampling implies faster learning while preserving geometric accuracy when compared with previous approaches. We also use the analytical derivatives of a neural implicit function to estimate the differential measures of the underlying point-sampled surface.
This work investigates the use of neural networks admitting high-order derivatives for modeling d... more This work investigates the use of neural networks admitting high-order derivatives for modeling dynamic variations of smooth implicit surfaces. For this purpose, it extends the representation of differentiable neural implicit surfaces to higher dimensions, which opens up mechanisms that allow to exploit geometric transformations in many settings, from animation and surface evolution to shape morphing and design galleries. The problem is modeled by a $k$-parameter family of surfaces $S_c$, specified as a neural network function $f : \mathbb{R}^3 \times \mathbb{R}^k \rightarrow \mathbb{R}$, where $S_c$ is the zero-level set of the implicit function $f(\cdot, c) : \mathbb{R}^3 \rightarrow \mathbb{R} $, with $c \in \mathbb{R}^k$, with variations induced by the control variable $c$. In that context, restricted to each coordinate of $\mathbb{R}^k$, the underlying representation is a neural homotopy which is the solution of a general partial differential equation.
This series presents lectures on research and development in visual computing for an audience of ... more This series presents lectures on research and development in visual computing for an audience of professional developers, researchers, and advanced students. Topics of interest include computational photography, animation, visualization, special effects, game design, image techniques, computational geometry, modeling, rendering, and others of interest to the visual computing system developer or researcher.
A manifold is a topological space that is locally Euclidean. Manifolds are important because they... more A manifold is a topological space that is locally Euclidean. Manifolds are important because they arise naturally in a variety of mathematical and physical applications as global objects with simpler local structure. In this paper we propose a technique for immersive visualization of relevant three-dimensional manifolds in the context of the Geometrization conjecture. The algorithm generalizes traditional computer graphics ray tracing. To do so we use several related definitions and results dating back to the works of Poincaré, Thurston, and Perelman.
A couple of years after its launch, NVidia RTX is established as the standard low-level real-time... more A couple of years after its launch, NVidia RTX is established as the standard low-level real-time ray tracing platform. From the start, it came with support for both triangle and procedural primitives. However, the workflow to deal with each primitive type is different in essence. Every triangle geometry uses a built-in intersection shader, resulting in straightforward shader table creation and indexing, analogous to common practice geometry management in game engines. Conversely, procedural geometry applications use intersection shaders that can be as generic or specific as they demand. For example, an intersection shader can be reused by several different primitive types or can be very specialized to deal with a specific geometry. This flexibility imposes strict design choices regarding hit groups, acceleration structures, and shader tables. The result is the difficulty that current game, graphics, and scientific engines have to integrate RTX procedural geometry (or RTX at all) in...
Journal of the Brazilian Computer Society, Apr 1, 1997
We discuss the problem of adaptive polygonization of regular surfaces of the euclidean 3D space, ... more We discuss the problem of adaptive polygonization of regular surfaces of the euclidean 3D space, and present e ective algorithms for computing optimal polygonizations of surfaces described in parametric or implicit form.
Figure 1: The collecTable running on the iTable (a), with several fiducials over it. The digital ... more Figure 1: The collecTable running on the iTable (a), with several fiducials over it. The digital projections of the fiducials floating over the interface (b), each fiducial storing different music collections. The M-Cube interface for music data showing albums (c) and tracks (d).
The novelty of our proposal is the end-to-end solution to combine computer generated elements and... more The novelty of our proposal is the end-to-end solution to combine computer generated elements and captured panoramas. This framework supports productions specially aimed at spherical displays (e.g., fulldomes). Full panoramas are popular in the computer graphics industry. However their common usage on environment lighting and reflection maps are often restrict to conventional displays. With a keen eye in what may be the next trend in the filmmaking industry, we address the particularities of those productions, exploring a new representation of the space by storing the depth together with the light map, in a full panoramic light-depth map.
We show how to use affine arithmetic to represent a parametric curve with a strip tree. The requi... more We show how to use affine arithmetic to represent a parametric curve with a strip tree. The required bounding rectangles for pieces of the curve are computed by exploiting the linear correlation information given by affine arithmetic. As an application, we show how to compute approximate distance fields for parametric curves.
In this paper, we propose the use of traditional animations, heuristic behavior and reinforcement... more In this paper, we propose the use of traditional animations, heuristic behavior and reinforcement learning in the creation of intelligent characters for computational media. The traditional animation and heuristic gives artistic control over the behavior while the reinforcement learning adds generalization. The use case presented is a dog character with a high-level controller in a 3D environment which is built around the desired behaviors to be learned, such as fetching an item. As the development of the environment is the key for learning, further analysis is conducted of how to build those learning environments, the effects of environment and agent modeling choices, training procedures and generalization of the learned behavior. This analysis builds insight of the aforementioned factors and may serve as guide in the development of environments in general.
Constructing high-quality generative models for 3D shapes is a fundamental task in computer visio... more Constructing high-quality generative models for 3D shapes is a fundamental task in computer vision with diverse applications in geometry processing, engineering, and design. Despite the recent progress in deep generative modelling, synthesis of finely detailed 3D surfaces, such as high-resolution point clouds, from scratch has not been achieved with existing approaches. In this work, we propose to employ the latent-space Laplacian pyramid representation within a hierarchical generative model for 3D point clouds. We combine the recently proposed latent-space GAN and Laplacian GAN architectures to form a multi-scale model capable of generating 3D point clouds at increasing levels of detail. Our evaluation demonstrates that our model outperforms the existing generative models for 3D point clouds.
Figure 1. Overview of our method: from a 2D photo and their corresponding facial landmarks (a)-(b... more Figure 1. Overview of our method: from a 2D photo and their corresponding facial landmarks (a)-(b), the facial texture data is extracted by successive 2D triangular subdivisions (c)-(d), producing a new 3D face model (e)-(f).
We introduce a novel framework for automatic 3D facial expression analysis in videos. Preliminary... more We introduce a novel framework for automatic 3D facial expression analysis in videos. Preliminary results demonstrate editing facial expression with facial expression recognition. We first build a 3D expression database to learn the expression space of a human face. The real-time 3D video data were captured by a camera/projector scanning system. From this database, we extract the geometry deformation independent of pose and illumination changes. All possible facial deformations of an individual make a nonlinear manifold embedded in a high dimensional space. To combine the manifolds of different subjects that vary significantly and are usually hard to align, we transfer the facial deformations in all training videos to one standard model. Lipschitz embedding embeds the normalized deformation of the standard model in a low dimensional generalized manifold. We learn a probabilistic expression model on the generalized manifold. To edit a facial expression of a new subject in 3D videos, the system searches over this generalized manifold for optimal replacement with the 'target' expression, which will be blended with the deformation in the previous frames to synthesize images of the new expression with the current head pose. Experimental results show that our method works effectively.
Region-based approaches to cel painting typically use shape similarity and topology relations bet... more Region-based approaches to cel painting typically use shape similarity and topology relations between regions of consecutive animation frames. This paper presents a new colorization algorithm based on topological differences defined over a hierarchical graph of adjacent regions, which allows an almost full automatic colorization process. Also this paper discusses other attributes that improve the solution of the image association problem.
In this paper we introduce an unified framework for basic operations on combinatorial 2manifolds ... more In this paper we introduce an unified framework for basic operations on combinatorial 2manifolds with or without boundary. We show that there are two kinds of primitive operators on the underlying meshes: operators which change the topological characteristic of the mesh and operators which just modify its combinatorial structure. We present such operators and demonstrate that they provide a complete set of elementary operations for mesh modification. We also give a description of the algorithms and data structures for an efficient implementation of these operators.
2022 35th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)
We present MR-Net, a general architecture for multiresolution neural networks, and a framework fo... more We present MR-Net, a general architecture for multiresolution neural networks, and a framework for imaging applications based on this architecture. Our coordinate-based networks are continuous both in space and in scale as they are composed of multiple stages that progressively add finer details. Besides that, they are a compact and efficient representation. We show examples of multiresolution image representation and applications to texture magnification, minification, and antialiasing. This document is the extended version of the paper [PNS + 22]. It includes additional material that would not fit the page limitations of the conference track for publication.
We introduce a neural implicit framework that exploits the differentiable properties of neural ne... more We introduce a neural implicit framework that exploits the differentiable properties of neural networks and the discrete geometry of point-sampled surfaces to approximate them as the level sets of neural implicit functions. To train a neural implicit function, we propose a loss functional that approximates a signed distance function, and allows terms with high-order derivatives, such as the alignment between the principal directions of curvature, to learn more geometric details. During training, we consider a non-uniform sampling strategy based on the curvatures of the point-sampled surface to prioritize points with more geometric details. This sampling implies faster learning while preserving geometric accuracy when compared with previous approaches. We also use the analytical derivatives of a neural implicit function to estimate the differential measures of the underlying point-sampled surface.
This work investigates the use of neural networks admitting high-order derivatives for modeling d... more This work investigates the use of neural networks admitting high-order derivatives for modeling dynamic variations of smooth implicit surfaces. For this purpose, it extends the representation of differentiable neural implicit surfaces to higher dimensions, which opens up mechanisms that allow to exploit geometric transformations in many settings, from animation and surface evolution to shape morphing and design galleries. The problem is modeled by a $k$-parameter family of surfaces $S_c$, specified as a neural network function $f : \mathbb{R}^3 \times \mathbb{R}^k \rightarrow \mathbb{R}$, where $S_c$ is the zero-level set of the implicit function $f(\cdot, c) : \mathbb{R}^3 \rightarrow \mathbb{R} $, with $c \in \mathbb{R}^k$, with variations induced by the control variable $c$. In that context, restricted to each coordinate of $\mathbb{R}^k$, the underlying representation is a neural homotopy which is the solution of a general partial differential equation.
This series presents lectures on research and development in visual computing for an audience of ... more This series presents lectures on research and development in visual computing for an audience of professional developers, researchers, and advanced students. Topics of interest include computational photography, animation, visualization, special effects, game design, image techniques, computational geometry, modeling, rendering, and others of interest to the visual computing system developer or researcher.
A manifold is a topological space that is locally Euclidean. Manifolds are important because they... more A manifold is a topological space that is locally Euclidean. Manifolds are important because they arise naturally in a variety of mathematical and physical applications as global objects with simpler local structure. In this paper we propose a technique for immersive visualization of relevant three-dimensional manifolds in the context of the Geometrization conjecture. The algorithm generalizes traditional computer graphics ray tracing. To do so we use several related definitions and results dating back to the works of Poincaré, Thurston, and Perelman.
A couple of years after its launch, NVidia RTX is established as the standard low-level real-time... more A couple of years after its launch, NVidia RTX is established as the standard low-level real-time ray tracing platform. From the start, it came with support for both triangle and procedural primitives. However, the workflow to deal with each primitive type is different in essence. Every triangle geometry uses a built-in intersection shader, resulting in straightforward shader table creation and indexing, analogous to common practice geometry management in game engines. Conversely, procedural geometry applications use intersection shaders that can be as generic or specific as they demand. For example, an intersection shader can be reused by several different primitive types or can be very specialized to deal with a specific geometry. This flexibility imposes strict design choices regarding hit groups, acceleration structures, and shader tables. The result is the difficulty that current game, graphics, and scientific engines have to integrate RTX procedural geometry (or RTX at all) in...
Uploads
Papers by Luiz Velho