Fast Uncertainty-driven Large-scale Volume Feature Extraction on Desktop PCs

Many of the current feature extraction techniques in large-scale application are designed around distributed environments. The ability to overcome the memory and bandwidth limitations of desktop PCs can broaden their usability towards large-scale applications. In this work, we present a new hybrid feature extraction technique which combines GPU-accelerated clustering with the multi-resolution advantages of supervoxels in order to handle large-scale datasets on standard desktop PCs. Furthermore, this is paired with a user-driven uncertainty-based refinement approach to enhance extraction results into a desired level of detail. We demonstrate the effectiveness and interactivity of this technique using a number of application specific examples utilizing large-scale volumetric datasets.


Franz Sauer is a Postdoctoral Scholar at the University of California, Davis, studying computer science and scientific visualization under Kwan-Liu Ma. His research interests include data visualization, large-scale scientific simulations, computer graphics, and physics. Before joining the ViDi group, Franz received a B.S. in physics from the California Institute of Technology.

Jinrong Xie is a PhD candidate in Computer Science at University of California, Davis. His research interests mainly include scientific visualization, large scale parallel graphics rendering and data analytics. He is working with Professor Kwan-liu Ma in the VIDI (Visualization and Interface Design Innovation) research group. Before joining UCDavis, he was a master student in the College of Computer Science and Technology, Zhejiang University. He holds a B.S. magna cum laude in the Department of Computer Science and Engineering of Shanghai Jiaotong University.