JASS 2004: Course 3 Reading List

Note: The citations given on this page are a mere starting point for your own literature scanning activities.

In this course, we will first discuss the state of the art in the seemingly unrelated areas of Augmented Reality and Ubiquitous Computing (UbiComp). The goal of the concept of Ubiquitous Tracking is to enable intelligent UbiComp environments to deliver positional data performant enough to enable Augmented Reality based user interfaces. The main problem with this combination is a unified treatment of tracking devices. To understand the problem, we will discuss the relevant mathematical concepts in the second part of the course. Once this is done, we can proceed with the thirs part, a discussion of the concepts and potential application areas of ubiquitous tracking.

Part 1: Augmented Reality and Ubiquitous Computing

Augmented Reality: Overview and Systems (Markus Geipel)

This presentation should give an overview of state-of-the-art Augmented Reality systems and technologies. It will explain the general processing pipeline of AR systems and highlight the key subcomponents of every AR system: tracking, 3D scene generation and multimodal user interaction. This talk will present a selection of AR systems, ranging from the Boeing Wire Assembly demo to current AR software frameworks. It will try to identify reoccurring problems and the evolution of ideas.

Ubiquitous and Context Aware Computing: Overview and Systems (Simon Bichler)

Ubiquitous Computing (UbiComp) is a new paradigm that aims at making computer systems invisibly interwoven with the user's environment. This talk will give an overview of the first ideas of UbiComp and applications that have been implemented to date. Context-aware systems, especially those that are mobile, are a fairly new field, but several successful systems have already been built. All of these are heavily under development and are used as research platforms. This presentation should give an overview of context awareness as a term, existing systems and present a few in more detail, showing goals, approach and current status.

Part 2: The Mathematics of Large-Scale Tracking

Overview and Mathematics of Tracking Technologies (Christoph Krautz)

Tracking is the process of repeatedly determining the position and orientation of objects or people. It is one of the core technologies of Augmented Reality and is at the center of this course's attention. In this talk, an overview of existing tracking technologies should be given, along with an intense discussion of advantages and drawbacks of different technologies. The representation of position and orientation in three-dimensional space and reoccuring problems therein will be discussed.

 

Sensor Fusion Systems: Overview and Mathematics (Björn Griesbach)

Sensor Fusion Systems: Overview and Mathematics (Björn Griesbach)

As will have been seen in the last talk, no single tracking technology is sufficient for all needs. The idea of combining multiple sensors to get a result that is better than the sum of the parts will be researched in this presentation. For this purpose, multiple multi sensor systems will be presented, with a discussion of commonalities and differences. In addition, the mathematics of two often used sensor fusion techniques, namely Kalman filters and particle filters, will be discussed.

 

The Mathematics of (Auto-)Calibrating AR Systems (Christian Wachinger & Benjamin Fingerle)

In AR setups, some relationships between objects and properties of cameras, displays etc. have to be determined before the system can be used. The mathematics to do so will be discussed in this talk. Additionally, this presentation gives an overview of research results aimed at automatic calibration of AR setups. Calibrating large setups without the user's intervention will be a key requirement to make the vision of Ubiquitous Tracking work.

Part 3: Combining Augmented Reality with Ubiquitous Computing

Foundations of Ubiquitous Tracking (Christian Wachinger & Benjamin Fingerle)

Ubiquitous Tracking aims at combining UbiComp environments with Augmented Reality interaction techniques. For this purpose, a concept for the dynamic integration of multiple sensors has to be provided that allows ad hoc integration of mobile users' tracking setups in intelligent environments. This talk will present some recent research results on that topic.

  -- MartinWagner - 25 Feb 2004