Publications

Memory Efficient Acceleration Structures and Techniques for CPU-based Volume Raycasting of Large Data

S. Grimm, S. Bruckner, A. Kanitsar, and M. E. Gröller

Abstract

Most CPU-based volume raycasting approaches achieve high performance by advanced memory layouts, space subdivision, and excessive pre-computing. Such approaches typically need an enormous amount of memory. They are limited to sizes which do not satisfy the medical data used in daily clinical routine. We present a new volume raycasting approach based on image-ordered raycasting with object-ordered processing, which is able to perform high-quality rendering of very large medical data in real-time on commodity computers. For large medical data such as computed tomographic (CT) angiography run-offs (512x512x1202) we achieve rendering times up to 2.5 fps on a commodity notebook. We achieve this by introducing a memory efficient acceleration technique for on-the-fly gradient estimation and a memory efficient hybrid removal and skipping technique of transparent regions. We employ quantized binary histograms, granular resolution octrees, and a cell invisibility cache. These acceleration structures require just a small extra storage of approximately 10%.

S. Grimm, S. Bruckner, A. Kanitsar, and M. E. Gröller, "Memory Efficient Acceleration Structures and Techniques for CPU-based Volume Raycasting of Large Data," in Proceedings of IEEE VolVis 2004, 2004, p. 1–8. doi:10.1109/SVVG.2004.8
[BibTeX]

Most CPU-based volume raycasting approaches achieve high performance by advanced memory layouts, space subdivision, and excessive pre-computing. Such approaches typically need an enormous amount of memory. They are limited to sizes which do not satisfy the medical data used in daily clinical routine. We present a new volume raycasting approach based on image-ordered raycasting with object-ordered processing, which is able to perform high-quality rendering of very large medical data in real-time on commodity computers. For large medical data such as computed tomographic (CT) angiography run-offs (512x512x1202) we achieve rendering times up to 2.5 fps on a commodity notebook. We achieve this by introducing a memory efficient acceleration technique for on-the-fly gradient estimation and a memory efficient hybrid removal and skipping technique of transparent regions. We employ quantized binary histograms, granular resolution octrees, and a cell invisibility cache. These acceleration structures require just a small extra storage of approximately 10%.
@INPROCEEDINGS {Grimm-2004-MEA,
author = "S{\"o}ren Grimm and Stefan Bruckner and Armin Kanitsar and Meister Eduard Gr{\"o}ller",
title = "Memory Efficient Acceleration Structures and Techniques for {CPU}-based Volume Raycasting of Large Data",
booktitle = "Proceedings of IEEE VolVis 2004",
year = "2004",
editor = "D. Silver, T. Ertl, C. Silva",
pages = "1--8",
month = "oct",
abstract = "Most CPU-based volume raycasting approaches achieve high performance  by advanced memory layouts, space subdivision, and excessive pre-computing.  Such approaches typically need an enormous amount of memory. They  are limited to sizes which do not satisfy the medical data used in  daily clinical routine. We present a new volume raycasting approach  based on image-ordered raycasting with object-ordered processing,  which is able to perform high-quality rendering of very large medical  data in real-time on commodity computers. For large medical data  such as computed tomographic (CT) angiography run-offs (512x512x1202)  we achieve rendering times up to 2.5 fps on a commodity notebook.  We achieve this by introducing a memory efficient acceleration technique  for on-the-fly gradient estimation and a memory efficient hybrid  removal and skipping technique of transparent regions. We employ  quantized binary histograms, granular resolution octrees, and a cell  invisibility cache. These acceleration structures require just a  small extra storage of approximately 10%.",
pdf = "pdfs/Grimm-2004-MEA.pdf",
images = "images/Grimm-2004-MEA.jpg",
thumbnails = "images/Grimm-2004-MEA.png",
youtube = "https://www.youtube.com/watch?v=WK9DJ6Dyrx4,https://www.youtube.com/watch?v=iYz5VYHMd9U,https://www.youtube.com/watch?v=UdtaaENWs7M",
affiliation = "tuwien",
doi = "10.1109/SVVG.2004.8",
isbn = "0-7803-8781-3",
keywords = "volume rendering, acceleration, large data",
url = "//www.cg.tuwien.ac.at/research/publications/2004/grimm-2004-memory/"
}
projectidprojectid

Media

Downloads

Full paper [PDF]