Trajectory Density Projection for Vector Field Visualization

Abstract

Trajectory visualization is an important tool to capture intrinsic characteristics of vector fields. However, often this class of geometric visualization suffers from visual clutter due to the number of curves that occlude each other and may cover relevant features. Attempts trying to avoid this effect often require complex pre/post-processing for seeding, clustering, or filtering of relevant lines. In this work we present a simple, yet effective technique for rendering large amounts of trajectories, which highlights features of their projected density. The technique exploits capabilities of modern graphics hardware and avoids explicit feature extraction. We propose efficient schemes for its computation, provide suggestions for sensible parameter setups, and discuss important implementation aspects. In addition, we demonstrate the effectiveness of our approach in several application scenarios.