With the increasing power of graphics hardware and the maturity of rendering algorithms, the ability to generate useful visualizations is now more often limited by human interaction with the system. Sophisticated visualization tasks and algorithms require the mastery of the details of the algorithm, the properties of the data, and the capabilities of the hardware. These hurdles often discourage those most knowledgeable of the underlying problem from driving the visual exploration process. As a result, the potential of the visualization is limited and the extent of scientific discovery may be reduced.
Our user interface research addresses this urgent need for innovation for demanding data visualization tasks.
Many new data reduction and advanced rendering methods have been invented for visualizing large, complex time-varying volume data sets. An equally important but often neglected aspect of a visualization solution is the accompanying interface through which the user makes, view, and manipulate visualizations. A carefully designed interface can make the exploration of large, complex data an easier job. The interface should abstract the complexity of visualization algorithms from the user, and display information in different but tightly coupled spaces to facilitate analysis and enable discovery.
An Interactive Interface for Visualizing Time-Varying Multivariate Volume Data. Hiroshi Akiba and Kwan-Liu Ma. In Proceedings of EuroVis 2007.
Layout of Multiple Views for Volume Visualization: A User Study. Daniel Lewis, Steve Haroz, and Kwan-Liu Ma. Proceedings of International Symposium on Visual Computing, November 6-8, 2006, pp. 215-226.
Simultaneous Classification of Time-Varying Volume Data Based on Time Histograms. Hiroshi Akiba, Nathan Fout, and Kwan-Liu Ma. In Proceedings of EuroVis 2006.
As both the scale and complexity of data analysis tasks continue to increase, various information about data exploration should be shared and reused to leverage the knowledge and experience scientists gain from data visualization. Essentially, we need to coherently manage, represent, and share information about both the visualization process and results (images and insights). Our approach to this problem is to encapsulate all information with a powerful visual interface that can help scientists keep track of their visualization experience and findings, use it to generate new visualizations, and share it with others. The concept of "visualizing visualizations" is introduced which could revolutionize the traditional paradigm for data visualization.
Visualizing Visualizations: User Interfaces for Management and Exploration Scientific Visualization Data, Kwan-Liu Ma, IEEE Computer Graphics and Applications, Vol. 20, No. 5, September/October 2000, pp. 16-19. [pdf]
A Sparedsheet interface presents the user visualization results organized in a tabular fashion. Data exploration becomes slicing through a multidimensional parameters space. Visualizations are refined by The user can operate on a whole row or column of the current table to refine visualizations.
A Model for the Visualization Exploration Process, T.J. Jankun-Kelly, Kwan-Liu Ma, and Michael Gertz, in Proceedings of IEEE Visualization 2002 Conference. [pdf]
Visualization Exploration and Encapsulation via a Spreadsheet-Like Interface, T.J. Jankun-Kelly and K.-L. Ma, IEEE Transactions on Visualization and Computer Graphics, 7(3), July-September 2001, pp 275-287. [pdf]
A Spreadsheet Interface for Volume Visualization, T.J. Jankun-Kelly and Kwan-Liu Ma, in Proceedings of the IEEE Visualization 2000 Conference. [pdf]
Image Graphs captures both the process and results of visualization, and can thus be reused and shared. Making new visualizations becomes operating on image graphs. It is also possible to optimize a visualization process by analyzing the image graphs.
Image Graphs - A Novel Interface for Visual Data Exploration, Kwan-Liu Ma, in Proceedings of the IEEE Visualization 1999 Conference. [pdf]
A Graph Based Interface for Representing Volume Visualization Results", James Patten and Kwan-Liu Ma, in Proceedings of Graphics Interface '98 Conference, 1998, pp. 117-124.
We develop intelligent visual interfaces which integrate machine learning into the data visualization process because many tedious visualization tasks such as segmentation, feature extraction and tracking can be effectively "learned" and performed by the computer, leaving the user free to concentrate on data understanding through an simple, intuitive user interface.
Intelligent Feature Extraction and Tracking for Large-Scale 4D Flow Simulations. Fan-Yin Tzeng and Kwan-Liu Ma. In Proceedings of Supercomputing 2005 Conference.
An Intelligent System Approach to Higher-Dimensional Classification of Volume Data. Fan-Yin Tzeng, Eric Lum, and Kwan-Liu Ma. IEEE Transactions on Visualization and Graphics, Volume 11, Number 3, May/June 2005, pp. 273-284.
A Cluster-Space Visual Interface for Arbitrary Dimensional Classification of Volume Data, Fan-Yin Tzeng and Kwan-Liu Ma. In Proceedings of VisSym 2004. [pdf]
A Novel Interface for Higher-Dimensional Classification of Volume Data, Fan-Yin Tzeng, Eric Lum, and Kwan-Liu Ma, in Proceedings of IEEE Visualization 2003 Conference [pdf]