And…Action! Setting the Scene for Accurate Visual CFD Comparisons Using Ray Tracing

Abstract

Increased graphical capabilities of contemporary computer hardware make ray tracing possible for a much wider range of applications. In science, and numerical fluid mechanics in particular, visual inspections still play a key role in both understanding flows, predicted by computational fluid dynamics, exhibiting features observable in real-life, such as interfaces or smoke, and when comparing such flows against experimental observations. Usually, little attention is paid to the visualisation itself, unless when the render is used solely for its eye-catching appearance. In this work, we argue that the use of ray tracing software can help make comparisons between computational and experimental fluid dynamics more robust and meaningful, and that, in some cases, it is even a necessity. Several visualisation problems which can be overcome through application of this methodology are discussed, and the use of ray tracing is exemplified for several common test cases in the maritime field. Using these examples the benefits of ray tracing are shown, and it is concluded that ray tracing can improve the reliability of scientific visual comparisons.

Multilevel Inverse Patchmatch Network with Local and Global Refinement for Underwater Stereo Matching

“…The cleanpass subset that we used consists of around 15K image pairs. The metric that we monitor is the end-point-error (EPE), calculated by 1 N ∑ i=1 |d i − d * i |, and the 1-pixel error (>1 px), as a percentage of the points with EPE larger than 1.…”

“…Stereo matching aims at recovering a scene's geometric information via stereo disparity estimation, which is a long-standing vision-based task for many underwater autonomous systems [1,2] and applications [3][4][5]. The early rectified stereo matching method focuses on finding the per-pixel disparity along the horizontal baseline in a rectified binocular image pair in a handcrafted manner.…”

Quantifying Uncertainties in Numerical Predictions of Dynamic Cavitation

“…A python module called pyTST which implements the "Transient Scanning Technique" presented in [65][66][67] is used to check whether the selected number of time step is sufficient or not. pyTST module allows to detect transient portion of a signal and measure the statistical uncertainty with that portion removed [68].…”