SCIENTIFIC COMPUTING AND IMAGING INSTITUTE
at the University of Utah

An internationally recognized leader in visualization, scientific computing, and image analysis

Vispack

Vispack is a C++ library that includes matrix, image, and volume objects. The library has functionality that makes it suitable for surface reconstruction work using the level sets method.


The Vispack Library

Introduction

Vispack is a C++ library developed for processing volumes images and surfaces. We will use it as the foundation for the projects in this course. Vispack relies on standard C++ features such as templates, inheritance, and the standard template library (STD). In the very least Vispack provides image readers and writers and a data structure for accessing image data. It also provides a number of image processing methods. Vispack does not provide any GUI interface to its functionality---it is strictly data structures, algorithms, and image IO.

Vispack Source

As with many open source packages, it has documentation, but it is not complete. You will learn vispack by reading the documentation, and by reading the source code. The vispack source code comes as a compressed tar file.

Now you can get the vispack source using CVS. On linux or Cygwin, you can check out the vispack as follows:
First, set an environment variable:
setenv CVS_RSH ssh
Then
cvs -d :ext:your-cs-login @ shell.cs.utah.edu:/uusoc/res/image/CVS checkout vispack

To update your local vispack
1. cd vispack
2. cvs update -d

or cvs -d :ext:your-cs-login @ famine.cs.utah.edu:/uusoc/res/image/CVS update
The systen then will ask for your password for your CS account.
Since we need to update Vispack from time to time, you are strongly recommended to check out vispack from CVS, which will save you time to get the latest version of Vispack.

CMake

Vispack has been compiled and tested on several different architectures including Mac OSX, SGI-IRIX, Linux, and Cygwin. It makes use of a build tool called CMake, which you will need to install. You can download CMake here .

Some Instructions

You should build vispack in a separate directory from the source. In this way we can update the source throughout the course as necessary. This is easy with Cmake. You simply create a sister directory to vispack, such as vispack-build. Change directories to vispack-build and then type "cmake ../vispack". You can configure cmake variables with "ccmake". You should make a soft link from vispack/apps/myapps to the directory where you will put your application directories. A CMakeList.txt file in that directory will indicate to Cmake that it should make your application subdirectories. By using this soft link we can later replace vispack without disturbing your applications.

Acknowledgements

The development of Vispack was supported by the Office of Naval Research under grant #N00014-01-10033 and the National Science Foundation under grant #CCR0092065.