Selected Tutorials
Gladimir Baranoski, T1: Hyperspectral Modeling of Material Appearance: General Framework, Challenges and Prospects.
The main purpose of this tutorial is to address theoretical and practical issues involved in the development of predictive material appearance models aimed at interdisciplinary applications within and outside the visible spectral domain. The tutorial begins by providing a concise review of relevant biophysical background, followed by a discussion on the key concept of predictability. The tutorial continues by examining the specific constraints and pitfalls found in each of the key stages of the model development framework, namely data collection, design and evaluation, and discussing alternatives to improve the fidelity of the entire process. Although predictive material appearance models developed by computer graphics researchers are usually aimed realistic image synthesis applications, they also provide valuable support for a myriad of advanced investigations in related areas, such as computer vision, image processing and pattern recognition, which rely on the accurate analysis and interpretation of material appearance attributes in the hyperspectral domain. In fact, their scope of contributions goes beyond the realm of traditional computer science applications. For example, predictive light transport simulations, which are essential for the development of these models, are also regularly being used by physical and life science researchers to understand and predict material appearance changes prompted by mechanisms which cannot be fully studied using standard ”wet” experimental procedures. Accordingly, the proposed tutorial closes with a detailed examination of recent examples of hyperspectral material appearance models which have been employed in such synergistic research efforts and in silico investigations.
Andreia Formico, T2: Desenvolvimento de Aplicações Gráficas Interativas com o Unreal Engine 4.
Neste tutorial de 3 horas, nível elementar, iremos apresentar os conceitos básicos para o desenvolvimento de aplicações gráficas interativas em tempo real, usando a Unreal Engine 4, a qual teve recentemente o seu código-fonte disponibilizado pela Epic Games. De forma geral, descreveremos a arquitetura da ferramenta e as principais classes que devem ser implementadas para iniciar o desenvolvimento de projetos na Unreal, nao são de forma teórica, mas também através de exemplos práticos. Usaremos o sistema de script visual da engine, o Blueprint Visual Scripting System, o qual funciona a partir de uma mecânica baseada em nós de controle para o desenvolvimentos de todos os elementos interativos do projeto (regras de jogo, jogadores, câmeras, entrada de dados e assets). Inicialmente, detalharemos como desenvolver a base para o comportamento de movimentação de um personagem em primeira pessoa (FPS), assim como a construção dos diversos elementos e materiais existentes no cenário. Em seguida, apresentaremos o procedimento para a criação de mecânicas de interação entre o usuário e os diferentes objetos e elementos do cenário, incluindo a utilização do editor de partículas da engine, o Cascade Particle Editor. Por fim, geraremos uma interface gráfica para a apresentação de dados e informações básicas do jogo (vida, munição, etc.), a partir da ferramenta específica provida pela Unreal, a Unreal Motion Graphics (UMG).
Esteban Clua, T3: Programming in CUDA for Kepler and Maxwell Architecture.
Since the first version of CUDA was launch, many improvements were made in GPU computing. Every new CUDA version included important novel features, turning this architecture more and more closely related to a typical parallel High Performance Language. This tutorial will teach how to program in GPUs using the latest CUDA versions, including concepts of dynamics parallelism, unified memory and concurrent kernels.
Tácito Tiburtino, T4: Visual Analysis using Multidimensional Projections.
This tutorial aims to present an overview of visual data analysis using multidmensional projections. The course highlights the process of producing effective visualizations using multidimensional techniques, taking users needs into account, by detailing the state-of-the-art techniques and tools. In order to provide practical experience, visualization procedures in specific case studies will be addressed. No previous experience is required in information or scientific visualization.
Fillipe Souza, T5: A Gentle Introduction to General Pattern Theory and its Application to Video Event Analysis.
In this tutorial, we will introduce basic concepts of General Pattern Theory (GPT) and demonstrate how to use its mathematical objects to represent and perform probabilistic inference of semantic video events. Understanding semantics of video events at descriptive level is desirable and important for numerous computer vision-based applications; however, this task goes beyond a simple classification problem. The goal is to be capable to inform where, when, why, how and what events are happening in a video given its feature observations and domain-specific knowledge information. The basic units of representation in GPT are called generators. Generators are connected through bond structures; thus, forming connected structures referred to as configurations. Configurations will typically represent complex structures of interest. The target set of generators forms a generator space (the pattern theoretic model chosen for a particular domain) that is dependent on domain-specific knowledge and rules of compatibility. The tutorial will comprise of two parts. The first part will be focused on introducing the basic principles and elements of Pattern Theory and give concrete examples of knowledge encoding. The second part will focus on explaining how to impose probabilistic superstructures on the algebraic model built on pattern theory. This will be followed by a discussion on the methods and algorithms that can be used to perform probabilistic inference with the pattern theoretic model. At the end of the second part, we will go over the details of how we have successfully applied pattern theory to the domain of semantic interpretation of video events.
For tutorials of previous years published by IEEE CPS please check:
For further questions, please, send e-mail to lrebouca@ufba.br, perfeuge@ufba.br or esdras@mat.ufc.br.