Informationen für

Logo VISUS



A Hybrid Physical/Device-Space Approach for Spatio-Temporally Coherent Interactive Texture Advection on Curved Surfaces

This web page contains additional material accompanying the paper "A Hybrid Physical/Device-Space Approach for Spatio-Temporally Coherent Interactive Texture Advection on Curved Surfaces".

Abstract

We propose a novel approach for a dense texture-based visualization of vector fields on curved surfaces. Our texture advection mechanism relies on a Lagrangian particle tracing that is simultaneously computed in the physical space of the object and in the device space of the image plane. This approach retains the benefits of previous image-space techniques, such as output sensitivity, independence from surface parameterization or mesh connectivity, and support for dynamic surfaces. At the same time, frame-to-frame coherence is achieved even when the camera position is changed, and potential inflow issues at silhouette lines are overcome. Noise input for texture advection is modeled as a solid 3D texture and constant spatial noise frequency on the image plane is achieved in a memory-efficient way by appropriately scaling the noise in physical space. For the final rendering, we propose color schemes to effectively combine the visualization of surface shape and flow. Hybrid physical/device-space texture advection can be efficiently implemented on GPUs and therefore supports interactive vector field visualization. Finally, we show some examples for typical applications in scientific visualization.

Publication Download

A PDF version can be found here.

Animated Visualization of an Automotive CFD Simulation

These two videos show the animated flow visualization of a CFD simulation data set from the automotive industry. The surface consists of 1,143,796 triangles and 597,069 vertices.

This video shows the animated flow visualization of a CFD simulation data set from the automotive industry. Here, the camera is fixed. We thank the BWM group for providing the underlying data set.

This video shows the animated flow visualization of a CFD simulation data set from the automotive industry. Here, the user moves the camera to explore the surface and the attached flow structures. We thank the BWM group for providing the underlying data set.

Comparision of Shading Approaches for Shape and Flow Visualization

These three videos compare different shading approaches for combined shape and flow visualization. The data set shows the results of a CFD simulation from the automotive industry. The surface consists of 1,143,796 triangles and 597,069 vertices.

This video shows a straightforward way of combining colors from the illumination with the structure from the flow texture: Gray-scale values are computed according to the diffuse illumination of the gray object surface and then modulated (i.e., multiplied) with the gray-scale values from the flow texture. This approach gives a good impression of both flow and object shape. However, both aspects are coded only by luminance variations and, therefore, other color dimensions are not used. We thank the BWM group for providing the underlying data set.

In this video, a flow texture is first mapped from gray-scale variations to hue variations between blue and yellow at (approximately) isoluminant levels. Afterwards, the transformed flow texture is modulated with an illuminated gray surface. We thank the BWM group for providing the underlying data set.

In this video, the roles of hue and luminance are exchanged. The LIC-like texture is represented by luminance variations and the surface by hue variations. Cool/warm shading supports the recognition of shape at nearly isoluminant colors. Here, cool/warm shading with yellowish and bluish colors is modulated with the gray-scale LIC-like texture. We thank the BWM group for providing the underlying data set.

Low Poygon Count Models

These four videos show simple surfaces with a low polygon count. The goal is to illustrate the behavior of our texture advection scheme at object boundaries, silhouette lines, and flow discontinuities.

This video shows a non-animated flow on a teapot. Please focus on the clear separation of the flow on neighboring foreground and background regions.

This video shows an animated flow on a teapot. Please focus on the clear separation of the flow on neighboring foreground and background regions.

This video shows an animated flow on a torus. The original 3D flow shows downwards along the vertical axis. Here, the vector field is normalized to unit length after the non-tangential (i.e., normal) component was removed. Caused by the removal of the normal vector component, the tangential flow field has a discontinuity at the topmost and bottommost parts of the torus. The visualization algorithm can handle these regions.

This video shows an animated flow on a torus. The original 3D flow shows downwards along the vertical axis. Here, the vector field is not normalized to unit length. Caused by the removal of the normal vector component, the tangential flow has zero magnitude at the topmost and bottommost parts of the torus.

HLSL Code

Our implementation of the texture-based surface flow visualization exploits the functionality and speed of graphics hardware by making extensive use of fragment programs. The implementation is based on C++ and DirectX 9. The fragment programs are formulated as HLSL shader programs. GPU programming is strictly separated from CPU programming by using effect files (fx files).

The part of the fx file that implements the relevant aspects of particle tracing can be downloaded.