-- CS590G (Fall 2004) --

Capturing, Modeling, Rendering 3D Structures

Have you ever wondered how to create models of 3D objects? Have you ever wanted to create a model of an entire room, floor, or building? Have you ever wanted to add real-world environments and objects to your games and virtual worlds? If so, this is the course for you!!!

           

Which of these two images represent the environment quality you'd rather see in your virtual world or game?

       

How do you think we should capture objects and the environment?

 

              

How do we use tools such as these to capture, model, and render 3D structures?

1. Overview

Summary: the objective of this course/seminar is to understand the fundamental problems and challenges encountered when capturing, modeling, and rendering 3D structures and objects. The course covers several subjects within computer graphics and computer science so as to provide to the student a full understanding of the capture/model/render pipeline. From this understanding and cross-fertilization of ideas, it is expected that students will in the future be able to develop new and improved approaches. The course is divided into three parts. The first part describes several capture devices, including standard field-of-view cameras and omnidirectional cameras. The emphasis of this section of the course is on making the student aware of the different types of capture devices and the fundamental problems common to all. The second part of the course describes several methods for modeling and rendering captured data. These methods range from laser-scanning approaches, to standard CAD modeling, to image-based approaches. The purpose of this section of the course is on providing an understanding of the state-of-art of current modeling and rendering research, with an emphasis on image-based approaches. The third part of the course describes several systems issues and display methodologies (e.g., methods for seeing objects in 3D). Throughout the course, demonstrations and videos will be presented.

Prerequisites: the course is open to graduates and undergraduates but students are required to have previous programming experience and are recommended to have previous computer graphics experience, although this is not mandatory. All students should seek approval by instructor for taking the course.

Course work: there is no book for the course; course material is composed of research papers and classroom presentations. The course work is a trivial warm-up assignment, two programming assignments and a mini-project. Assignments must be done individually but the mini-project may be done in small groups (e.g., 2 students). In addition, students are expected to give a presentation of an existing research paper. The paper may be related to their mini-project and/or can be chosen from a provided list of papers. The classroom load towards the end of the course is less so as to allow students to work on their projects.

Classroom: Univ 217

Time: T/Th 10:30-11:45

Office hours: by appointment

Grading:

    Assignments #0, #1, and #2: 40% of grade (distributed roughly 5%, 15%, 20%)

    Mini-Projects: 40% of grade

    Paper presentation/Class Participation/Attendance: 20% of grade

    No mid-term exams, final exam, tests or quizzes.

    Extra credit items encouraged throughout course.

Late policy:

    Assignments due before class on due date.

    1st time late: up to one week, but instructor must be notified in advance -- no penalty.

    2nd and subsequent times: grade reduction of 15% per day.

    All assignments, mini-project, and presentations required by demo day at end of course.

 


 

2. Assignments

Assignment #0

Handout

Pictures:

     [PPM]        [PPM]       [PPM]

 

GLUI libraries and includes

GLUT libraries and includes

JPEG libraries and includes

 

Assignment #1

Handout

The calibration software: tsai.zip.

A JPEG to PPM image file converter: jpegtopnm.zip.

 

Assignment #2

Handout

The triangulation software is in triangulation.zip.

 

Final Project

Handout

 


3. Slides

Introduction

Pipeline, Capture Devices

Calibration

Omnidirectional cameras

Pose I

Pose II

Art Gallery

Aspect Graph

Computer Vision in a Nutshell

Image Morphing and View Morphing

Image Warping and View-Dependent Texture Mapping

IBR and Lightfields

Sea of Images

Plenoptic Stitching

Final Project