Team Calvi

Overall Objectives
Scientific Foundations
Application Domains
New Results
Contracts and Grants with Industry

Section: New Results

Keywords : Multi-GPU, Pre-integrated, Sort Last Volume.


Participants : Jean-Michel Dischler, St├ęphane Marchesin.

Higher order Pre-integrated Volume Rendering

In the field of Volume Rendering, pre-integration techniques for arbitrary transfer functions has certainly led to the most significant and convincing results on standard PC consumer graphics with regard to both quality and performance. By showing that the ideal scalar signal along the cast rays is better approximated by a succession of polynomial curves as opposed to linear segments, we have proposed a new method for pre-integrated volume rendering. This method is based on a second and third order polynomial interpolation of the scalar values, allowing it to converge more rapidly towards the integration of a volume reconstructed by a trilinear filter. Our approach manages to capture the smoothness of the volume's details without the need of further ray re-sampling, and consequently succeeds in reducing the visual artifacts in comparison to previous techniques. Our experiments showed that second order polynomials improve framerates and visual quality while third order does not provide significant further improvements, which is related to the fact that sample positions are equispaced for real-time rendering purposes. We published second order pre-integration in [35] .

Multi-Sort Last Volume Visualization

We conducted an experimental study of an inexpensive off-the-shelf sort-last volume visualization architecture based upon multiple GPUs and a single CPU. This study showed how to efficiently make use of this architecture to achieve high performance sort-last volume visualization of large datasets. We analyzed the bottlenecks of this architecture and compared it to a classical sort-last visualization system using a cluster of commodity machines interconnected by a gigabit Ethernet network. As a result of this experimental study we showed that this solution competes very well with a mid-sized PC cluster, while it significantly improves performance compared to a single standard PC. The results of our study have been published in [38] .


Logo Inria