Efficient software for programmable visual analysis using Morse-Smale complexes.

Nithin Shivashankar and Vijay Natarajan.
In Topological Methods in Data Analysis and Visualization IV.
Hamish Carr, Christoph Garth, and Tino Weinkauf (Eds.)
Springer-Verlag, Mathematics and Visualization Series, 2017, 317-331.

Abstract

The Morse-Smale complex is a topological data structure that represents the behavior of the gradient of an input scalar field. Recent years have witnessed a significant number of applications that use this data structure for visualization and analysis of data from various scientific domains. However, these applications have required significant expertise in the implementation of algorithms. This potentially makes such analysis inaccessible to a large audience. In this paper we present open source software modules for the computation, analysis, and visualization of scientific data using the Morse-Smale complex. The modules, named pymstri and pyms3d, are intended for domains represented using 2D triangle meshes and 3D structured grids respectively. The software is designed to significantly reduce the effort required to use Morse-Smale complex based analysis. Also, the software leverages modern multi-core CPU and GPU architectures for computational efficiency. We demonstrate the usefulness via a case study to visually analyze and interactively segment the eye of the Hurricane Isabel simulation dataset. In particular, we highlight the ability to couple the visual analysis and the computation with ParaView, a popular general purpose visualization tool. The code is is available at the project website here.

[PDF]