TUTORIALS (Selected Tutorials below)

The goal of the tutorials at SIBGRAPI/SIACG 2004 is to offer to the participants an opportunity to acquire technical knowledge in the areas of Computer Graphics and Image Processing, and/or applicability to other areas. Tutorial proposals will be judged by:

  • their relevance for the SIBGRAPI/SIACG 2004
  • the expected size of the audience and their potential to attract participants to the Symposium
  • the validity, usefulness, and structure of the contents
  • their originality and
  • the qualification of the instructors in the topic

  • Types of Tutorials

      The tutorials may be elementary, advanced, state-of-the-art or hands-on, and may take 3 or 6 hours, regardless of the level.

      • Elementary Tutorials: Tutorials in this category typically complement the basic undergraduate curriculum in computer science and help to attract students for graduate studies in CG & IP. The instructors should not assume that the audience has basic knowledge in CG & IP. Elementary tutorials should be presented and have handouts in Portuguese or Spanish.

      • Advanced Tutorials: Tutorials in this category are targetted to students, professionals and researchers in CG&IP that wish to know advanced techniques to use in their work Instructors may assume that the audience has basic notions of mathematics, numerical methods, programming, etc., and about CG & IP. Advanced tutorials may be presented in Portuguese, Spanish or English, with handouts in any of these languages.

      • State-of-the-art Tutorials: This type of tutorial is new for this year, and should cover recent developments in CG&PI that may have great impact in the future of the area. State-of-the-art tutorials may be presented in Portuguese, Spanish or English, with handouts in any of these languages.

      • Hands-on Tutorials: Tutorials in this category should be target towards undergradute computer science students covering basic topics in Computer Graphics or Image Processing. Some possible topics include but are not limited to: introduction to computer graphics using OpenGL, OpenGL advanced features, Web 3D using X3D, introduction to visualization, introduction to image processing using Matlab, Mathematica, Maple, IDL, ENVI, etc.; elements of mathematical morphology, and introduction to medical images. State-of-the-art tutorials may be presented in Portuguese, Spanish or English, with handouts in any of these languages.

      It may be useful to look at the tutorials presented in previous years: 2003, 2002, 2001, 2000 , 1999, 1998, 1997.

  • How to Submit

      Tutorial proposals should be submitted by e-mail to Manuel M. Oliveira, in plain text or in attached compressed Postscript or PDF files. Please, do not send Word documents.

      The proposal should contain the following information:

        Title

        Level (hands-on, elementary, advanced or state-of-the-art)

        Duration (3 or 6 hours)

        Abstract (half page)

        Motivation, target audience, and interest for the CG&IP community (up to 1 page)

        Summary, presenting an itemized list of the topics to be presented (up to 2 pages). For each topic, indicate the estimated duration, the basic and most relevant literature, and its subtopics.

        Curriculum Vitae of the instructors, with full name, address, e-mail, institution, education, publications, and experience in the topic of the tutorial.

        Material to be distributed to the participants, such as survey paper (recommended), slides, images, animations, books, etc. This material is optional at the submission, but its inclusion is strongly recommended as it allows a better assessment of the supporting material to be used during the course. For hands-on tutorials, the software to be used should be specified (licenses and installation are responsability of the intructors). See below.

        Presentation equipment required (for hands-on tutorials, specify computers configuration needed).

      The tutorial could be prepared either in English or Portuguese.

      The complete submission should be received by 30th May, 2004. Incomplete or late submissions will not be considered. All submissions will be promptly acknowledged. Accepted tutorials will be announced by 20th June, 2004

  • Printed Material

      Instructors of the tutorials selected for SIBGRAPI/SIACG 2004 should prepare written material to suport the lectures, in the form of a survey paper (preferably) and/or handouts. This material will be reproduced in black & white and distributed to the participants of the tutorial.

      • Survey paper is a paper having 15-30 pages, with an introduction, a structured presentation of the topic, and conclusions indicating trends, applications, and directions for future work. 

        Survey papers will be peer-reviewed and published in a special issue of RITA (Revista de Informática Teórica e Aplicada), a journal edited by the Informatics Institute of UFRGS. See the information for authors for instructions on how to format survey papers, including LaTeX and Word styles.

        Survey papers should be sent no later than 20th July, 2004 by e-mail to Manuel M. Oliveira, in attached compressed Postscript or PDF files. 
         

      • Handout is a copy of the slides or transparencies used during the presentation of the course at SIBGRAPI/SIACG 2004. Handouts are optional if a survey paper is prepared.

        The authors should provide a file with 4 slides per page in landscape. The pages should be numbered and contain the footer
        "SIBGRAPI/SIACG 2004 Tutorial". We recommend the use of font Times 14 pt, or bigger, for better legibility of slides and notes.

        Handouts should be sent no later than July 20th, 2004 by e-mail to Manuel M. Oliveira  in attached compressed Postscript or PDF files. Please, do not send Word documents.

      All printed material distributed to the participants of tutorials will be available on the SIBGRAPI/SIACG 2004 web site, as well as other material provided by the instructors such as images, animations, and other texts.

  • Support to Instructors

      A classroom for 60 students will be reserved for each course. Standard presentation equipment (overhead projector, 35 mm slide projector, video player, screen, PC, data show) will be available upon request.

  • Important Dates

      Tutorial proposals due: May 30, 2004.
      Selected tutorials announced: June 20, 2004.
      Survey papers due: July 20, 2004.
      Other printed material due: September 10, 2004.

  • Selected Tutorials

    • Tutorial #1

        Title: Perception for Visualization: From Design to Evaluation
        Authors: Haim Levkowitz and Maria Cristina Ferreira de Oliveira
        Affiliations: University of Massachusetts Lowell and ICMC - USP - São Carlos
        Level: Advanced
        Duration: 3 hours
        Abstract: What is the smallest sample I can show that will be perceived? What is the smallest sample I can show that will be perceived in color? Can I afford using image compression? If yes, how much and what kind? Should I use a grayscale or another color scale? How many gray levels do I absolutely need? What color scale should I use? How many bits for color do I need to have? Should I use 3D, stereo, texture, motion? If so what kinds? Should I add multiple views to my visualization? If so, how many? What about adding sonification? And, finally, Has my visualization been successful meeting its goals and needs? If you have ever designed a data visualization, you probably have asked yourself (and perhaps other people) some of these questions; at least you should have. Since visualization "consumers" are human beings, the answers to these questions can only come from a thorough analysis and understanding of human perceptual capabilities and limitations, combined with a visualization's goals and needs. This tutorial will teach you the basics of human perception and how to utilize them in the complete process of visualization: from initial design to final evaluation. While a major part of the tutorial will focus on visualization literally, we will also address similar principles when designing a sonification, or a combined visualization and sonification.
        Who should attend: Anybody engaged in the design, implementation, and evaluation of data visualizations.
        Prerequisites: Knowledge and understanding of basic computer graphics, imaging, and visualization concepts.
        What will you gain: Understanding of the complete processes of visualization and sonification, from design to evaluation, and how to incorporate perception into the process to maximize the visualizations impact and results.

    • Tutorial #2

        Title: Computer Vision and Graphics for Heritage Preservation and Digital Archaeology
        Authors: Luciano Silva, Olga R. P. Bellon, and Kim L. Boyer
        Affiliations: Universidade Federal do Paraná and The Ohio State University
        Level: State-of-the-Art
        Duration: 3 hours
        Abstract: The goal of this tutorial is to provide attendees with a survey of topics related to Heritage Preservation and Digital Archaeology, which are challenging and motivating subjects to both computer vision and graphics community. These issues have been gaining increasing attention and priority within the scientific scenario and among funding agencies and development organizations over the last years. Motivations to this work are the recent efforts in the digital preservation of cultural heritage objects and sites before degradation or damage caused by environmental factors or human development. One of the main focuses of these researches is the development of new techniques for realistic 3D model building from images, preserving as much information as possible. In this tutorial we address this problem and propose a robust approach expected to be useful for many other purposes. We intend to introduce and discuss several emerging topics in computer vision and graphics related to heritage preservation and digital archaeology while highlighting the major contributions and advances in these fields. Topics covered include range images, multiview registration, 3D modeling and representation, robust methods and current research challenges. Emphasis can be given to selected advanced topics depending on the interests of participants.

    • Tutorial #3

        Title: An Introduction to Light Interaction with Human Skin
        Authors: Gladimir V. Baranoski and Aravind Krishnaswamy
        Affiliations: Natural Phenomena Simulation Group, University of Waterloo
        Level: Advanced
        Duration: 3 hours
        Abstract: Despite the notable progress in physically-based rendering, there is still a long way to go before one can auto-matically generate predictable images of biological materials. This tutorial addresses an important topic in this area, namely the realistic rendering of skin, the largest organ in the human body. This tutorial begins with a review of relevant measurement procedures used in data collection and model validation, followed by a presentation of biological factors involved in the processes associated with light propagation in skin tissues, namely surface reflectance, subsurface reflectance and transmittance as well as the mechanisms of light absorption by natural pigments.
        These processes not only affect skin appearance, but also its health. Not surprisingly, there is a considerable amount of research on skin optics available in the biomedical literature as well as recent investigations in pattern recognition and colorimetry. The reflectance models developed in these fields are mostly aimed at the reproduction of skin spectral properties to determine the content and distribution of various substances, i.e., scattering properties affecting skin appearance are usually not addressed. In computer graphics, on the other hand, the focus has been on addressing these properties and developing scattering models to be used in image synthesis frameworks. These computer graphics models, however, incorporate algorithms and techniques used in the biomedical models. More recently, research efforts have been directed toward the development of com-prehensive models, which take as input biological and structural data and provide as output both spectral and scattering quantities for skin specimens.
        Taking into account these aspects, this tutorial provides a review of the most important modeling approaches used in biomedical applications followed by a more detailed description of current models used in image syn-thesis frameworks. The tutorial will conclude with a discussion of open problems and current trends in the predictive rendering of human skin.

    • Tutorial #4

        Title: Filtros Preditivos
        Author: Siome Klein Goldenstein
        Affiliation: UNICAMP
        Level: Advanced (in Portuguese)
        Duration: 4 hours
        Abstract: Filtros preditivos são um conjunto de técnicas para realizar estimação Bayesiana de estados. Estas técnicas são frequentemente utilizadas em visão computacional, controle e robótica, para rastreamento e observação. A primeira etapa para seu uso é a caracterização do sistema em questão. É necessário identificar o conjunto de variáveis essenciais para a descrição do estado desse sistema. Isto não é um problema simples, e também não possui solução única, exigindo portanto experiência e cuidado por parte do projetista. A segunda etapa consiste na modelage, da dinâmica deste sistema sem a presença de fatores externos. Ou seja, como as variaveis de estado mudam por conta própria. Há duas formas de representar isso: como equações diferencias (quando desejamos uma representação das variaveis de estado no tempo contínuo), ou como equações de diferença (quando estamos apenas interessados em uma amostragem discreta do tempo). Para a terceira etapa necessitamos descrever este sistema de forma Bayesiana. Os valores das variáveis de estado se tornal variáveis aleatórias, e a dinâmica passa a ser descrita como probabilidades condicionais. O processo de estimação Bayesiana dos estados é um processo indutivo. Partimos da representação do estado atual (prior), temos a dinâmica do sistema (descrita como uma probabilidade condicional), e coletamos uma observação. Utilizamos a Bayes para obter a distribuição da predição a partir do prior, e depois executamos uma fusão de dados com a informação obtida da observação. Todos os filtros preditivos fazem essencialmente a mesma coisa, diferindo apenas em suposições e técnicas para resolver cada etapa. O Filtro de Kalman é utilizado quando o sistema possui dinâmica e observações lineares, e variáveis de estado com distribuições Gaussianas. Quando o sistema é não linear, porém ainda deseja-se descrever as distribuições como Gaussianas, temos então o Filtro de Kalman Extendido. Quando as distribuições são não paramétricas, utiliza-se técnicas de amostragem, como nos métodos de Monte Carlo. São os casos do Filtro de Partículas e Filtro Unscented.

    • Tutorial #5

        Title: Introdução à programação com Java Advanced Imaging
        Author: Rafael Santos
        Affiliation:Instituto Brasileiro de Tecnologia Avançada (IBTA), São José dos Campos
        Level: Hands-on (in Portuguese)
        Duration: 4 hours
        Abstract: JAI (Java Advanced Imaging) é uma API (application programming interface), ou conjunto de classes, que permite a inclusão de uma ampla gama de rotinas de processamento de imagens em aplicações escritas em Java. JAI permite também o acesso paginado a imagens grandes (por exemplo, que não caberiam na memória principal do computador). O poder e a flexibilidade do JAI, aliados à facilidade de programação em Java, permitem a criação rápida de aplicações de processamento de imagens. Como vantagem adicional, tanto Java quanto JAI podem ser executados em praticamente qualquer plataforma, podem ser obtidos gratuitamente e sem restrições de distribuição, o que faz desta combinação uma opção atraente quando comparada com soluções comerciais de ambientes de desenvolvimento para processamento de imagens.
        Infelizmente, tutoriais básicos e práticos para programação usando o JAI inexistem, e muitos programadores novatos não têm como aprender os conceitos básicos de JAI para desenvolver seus projetos. Este tutorial tem como objetivo apresentar os conceitos básicos de JAI através de exemplos práticos e funcionais. Este tutorial apresentará os conceitos elementares de Java (somente o suficiente para que programadores em C, C++ ou Pascal compreendam as diferenças), os conceitos elementares do JAI e a implementação de rotinas conhecidas (filtros, conversores, classificadores, ferramentas de visualização etc.) em JAI.

    • Tutorial #6

        Title: Shader Programming: An Introduction Using the Effect Framework
        Author: Joern Loviscach
        Affiliation: Hochschule Bremen, Germany
        Level: Hands-on
        Duration: 4 hours
        Abstract: Current commodity graphics cards offer programmability through vertex shaders and pixel shaders to create special effects by deformation, lighting, tex-turing, etc. The Effect framework introduced by Microsoft and Nvidia allows to store shader program code, settings, and a limited graphical user interface within a single .fx text file. This supports sharing work between programmers who write the code and designers who manipulate the GUI elements. Furthermore, the Effect framework proves to be ideal for experimenting with shader programming-be it for learning purposes or for rapid prototyping of new algorithms. In this tutorial, we employ the Effect framework for an exploratory, hands-on approach, introducing first principles only as needed, not in advance. Simple shader programs are used to review basic 3D techniques such as homogeneous coordinates and the Phong shading model. Then we turn to basic deformation effects through vertex shaders and the use of texture maps as decals or reflected environments inside pixel shaders. To create bump mapping and related effects, tangent space coordinates and normal maps are introduced. Finally, we treat more complex effects such as anisotropic specular highlights, procedural tex-tures, and alpha blending.


  • Contact