3D visualization, of biomedical data in particular, is a complex process relying on sophisticated algorithms and techniques. There are some pure Java solutions available, for instance VolumeJ and SurfaceJ plugins from Michael Abramoff. However, there is no tool or library that can match versatility of VTK (suggestions of possibly overlooked options are very welcome, post comments at ij-plugins user's list or forum, optinally contact directly the project's admin).
The Visualization ToolKit (VTK) is an open source, freely available software system for 3D computer graphics, image processing, and visualization. VTK consists of a C++ class library, and several interpreted interface layers including Tcl/Tk, Java, and Python. VTK supports a wide variety of visualization algorithms including scalar, vector, tensor, texture, and volumetric methods; and advanced modeling techniques such as implicit modelling, polygon reduction, mesh smoothing, cutting, contouring, and Delaunay triangulation.
Rather than reinventing the wheel (a "pure Java" solution), the ij-VTK attempts to make the best use of already available and proven visualization solution. This may not satisfy needs of every ImageJ user, especially those on platforms that currently do not have complete VTK support for Java (e.g. Mac OS X 10.2 and earlier). However, reuse rather than redevelopment enables an optimal utilization of resources available to this project, to deliver new visualization functionality to many, if not most, ImageJ users.
Similar to ImageJ, VTK have very active user/developer community. It is foreseeable that with enough interest platforms like Mac OS X 10.2 will have its VTK support improved. This even may no longer be an issue in newer Mac OS and Java version, like Mac OS X 10.3 and upcoming Java 1.5 (tiger) with revamped support for native graphics.
ij-VTK consists of two main parts: ij-VTK Core Utilities and ij-VTK Plugins for ImageJ. The following diagram illustrates relationship between ij-VTK, VTK, ImageJ, and ij-plugins Toolkit (core library of the ij-plugins project):
VTK does contain wrappers that enable its use from Java. However, VTK uses event handling, error reporting, and display access that is not very familiar to Java developers. ij-VTK will contain support classes aimed at simplifying VTK use from Java and ImageJ.
ij-VTK does depend on operating system specific VTK's native libraries. "Standard" installation of VTK typically requires compilation of VTK C++ source code and a tricky setup of system variables.
ij-VTK aims at simplifying VTK setup by providing all required VTK binaries in an easy to install package customized for ImageJ users. The initial goal is to provide binaries for Linux and MS Windows. Volunteers are needed for creation, testing, and maintenance of binaries for other operating systems.
The Core Utilities will provide functionality aimed primarily at accessing VTK through paradigms similar to Java beans or Swing GUI library. For instance, VTK callback functions for event handling will be wrapped using event listener interfaces similar to those used by Swing components. Here is some of the functionality planned for ij-VTK Core Utilities:
Wrapper listener for events generated by individual VTK objects
Monitoring of progress of VTK pipeline execution - aggregation of progress generated by individual VTK objects
Data transfer between Java side and C++ side
VTK actor property editors
Top level monitoring of VTK errors and warnings (after release 1.0, requires custom JNI/C++ code)
Screen capture (after release 1.0)
VTK pipeline editors (after release 1.0)
All visual plugins will support interactive rotating, zooming, and panning of displayed 3D data and high resolution screen capture in TIFF format. Some of the planed plugins:
Volume rendering of 3D image data
Interactive editing of transfer function
Interactive volume clipping using re-sizable box widget
Surface rendering of 3D image data
Image filters for 2D and 3D processing (non-visual plugins)
Surface plots for 2D images (after release 1.0)
Color histograms and 3D color space visualization: RGB, CIE Lab, HSB (after release 1.0)