Improvement of particle-based volume rendering for visualizing irregular volume data sets

Abstract

We present a technique for previewing large-scale irregular volume datasets using an improved particle-based volume rendering (PBVR) technique. Volume rendering of irregular grid volume data is considerably more complicated than that of regular grid data, since the sampling and compositing processes, which must be done in visibility order, are not straightforward. In our original PBVR, rendering involves generating and projecting sub-pixel-size, opaque, and emissive particles without visibility ordering. To make it easier to preview large-scale irregular volume datasets, we improve our original PBVR technique in two respects. The first is that we exploit its scalability to develop a cell-bycell particle generation technique. The second is that we reduce the memory cost of the frame buffer using a pixel-by-pixel superimposing technique. To measure the effectiveness of our proposed method, we apply it to huge irregular volume datasets composed of 71 mega hexahedral cells or 1 giga tetrahedral cells. Keywords volume rendering, large-scale irregular volume datasets, tetrahedral cells, hexahedral cells. 1 INTRODUCTION Many volume rendering methods have been developed over the past two decades. In fact, the development of techniques for irregular volumes remains a challenging area in the visualization community. Irregular volume datasets consist mainly of scalar data defined on collections of irregularly ordered cells whose shapes are not necessarily orthogonal cubic. Data of this type can be found in the results of the finite element method (FEM) technique, which is widely used in computational mechanics. It is also becoming popular in computational fluid dynamics (CFD). High performance computing calculates a huge FEM model, which cannot reside in a single computational node and is distributed to multiple computational nodes. A user must be able to do a fast preview in order to evaluate whether the computation was successful. For previewing a volume dataset, a volume rendering technique is suitable, since it lets the user visualize the whole volume space. However, splatting requires visibility sorting, in which all volume cells need to be processed in advance at each viewing point. Our particle-based volume rendering (PBVR) technique is suitable for previewing a huge irregular volume dataset, since it requires no visibility sorting. The technique represents a given volume dataset as a set of particles that are emissive and opaque, based on a particle model derived from Sabella's density emitter model [1]. Once a volume dataset is converted to a set of particles, the rendering process is efficient, regardless of the visibility order. Scalability in the number of the volume cells is crucial to successful volume rendering of huge irregular volume datasets. We were therefore obliged to modify our original PBVR so that it could process an irregular volume dataset cell-by-cell. In this paper, we describe an improved PBVR that make it easier to preview a huge irregular volume dataset. One improvement is that it generates pa...

Parallel Caching for Scientific Visualization

“…We expect to deploy and further test our framework in this context. Future works also include extensions to integrate point based volume rendering [11], [12], pursuing the initial work performed with FlowVR [17]. Data sampling could be perfomed on the server side and the effective point based rendering on the client side.…”

“…This is very common for terrain rendering (mesh resolution is often also adapted according to the distance with the view point) [9], but has also been used for scientific visualization [10]. Sakamoto et al [11], [12] propose to combine progressive rendering and point based rendering for volume rendering. The data set is resampled at a given rate, enabling to control the image quality.…”

Spatio-temporal visualization of air–sea CO2 flux and carbon budget using volume rendering

“…Since Reeves (1983) presented particle systems for the volume rendering of fuzzy objects without smooth, well-defined and shiny surfaces, it has been widely used for the simulation of fluid, such as water, fire and smoke (Horvath and Geiger, 2009;Pfaff et al, 2010;Sakamoto et al, 2010;Yu and Turk, 2013). Voxels in particle systems are supposed to emit, absorb and scatter light.…”

A Versatile Optical Model for Hybrid Rendering of Volume Data

“…In object-ordered rendering techniques, sometimes particles can also be generated explicitly [4]. At an optical model level, it is basically the same as the continuous integration form.…”