Abstract
Dark matter simulations, performed using N-body methods with a finite set of tracer particles to discretize the initially uniform distribution of mass, are an invaluable method for exploring the formation of the universe. Definining a tetrahedral mesh in phase space-with the tracer particles at initialization serving as vertices-yields a more accurate density field. At later timesteps, the mesh self-intersects to an enormous degree, making pre-sorting impossible. Kaehler et al [2012] visualize the mesh using cell projection, but their method requires order-independent compositing, which limits its flexibility. Our work renders the mesh using state of the art order-independent transparency (OIT) techniques to composite fragments in correct depth order. This also allows us to render variables other than density, such as velocity. We implement a number of OIT optimizations to handle the high depth complexity (on the order of 107 depth layers for 2×109 particles) of the data. Our performance measurements show near-interactive framerates for our hybrid renderer despite the large number of depth layers.
| Original language | English (US) |
|---|---|
| Title of host publication | SA 2016 - SIGGRAPH ASIA 2016 Symposium on Visualization |
| Publisher | Association for Computing Machinery, Inc |
| ISBN (Electronic) | 9781450345477 |
| DOIs | |
| State | Published - Nov 28 2016 |
| Event | 2016 SIGGRAPH ASIA Symposium on Visualization, SA 2016 - Macau, China Duration: Dec 5 2016 → Dec 8 2016 |
Other
| Other | 2016 SIGGRAPH ASIA Symposium on Visualization, SA 2016 |
|---|---|
| Country | China |
| City | Macau |
| Period | 12/5/16 → 12/8/16 |
Fingerprint
Keywords
- GPU acceleration
- Intersecting mesh
- Order-independent transparency
- Scientific visualization
- Volume rendering
ASJC Scopus subject areas
- Computer Vision and Pattern Recognition
- Human-Computer Interaction
- Software
- Computer Graphics and Computer-Aided Design
Cite this
Volume rendering dark matter simulations using cell projection and order-independent transparency. / Igouchkine, Oleg; Leaf, Nick; Ma, Kwan-Liu.
SA 2016 - SIGGRAPH ASIA 2016 Symposium on Visualization. Association for Computing Machinery, Inc, 2016. a8.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Volume rendering dark matter simulations using cell projection and order-independent transparency
AU - Igouchkine, Oleg
AU - Leaf, Nick
AU - Ma, Kwan-Liu
PY - 2016/11/28
Y1 - 2016/11/28
N2 - Dark matter simulations, performed using N-body methods with a finite set of tracer particles to discretize the initially uniform distribution of mass, are an invaluable method for exploring the formation of the universe. Definining a tetrahedral mesh in phase space-with the tracer particles at initialization serving as vertices-yields a more accurate density field. At later timesteps, the mesh self-intersects to an enormous degree, making pre-sorting impossible. Kaehler et al [2012] visualize the mesh using cell projection, but their method requires order-independent compositing, which limits its flexibility. Our work renders the mesh using state of the art order-independent transparency (OIT) techniques to composite fragments in correct depth order. This also allows us to render variables other than density, such as velocity. We implement a number of OIT optimizations to handle the high depth complexity (on the order of 107 depth layers for 2×109 particles) of the data. Our performance measurements show near-interactive framerates for our hybrid renderer despite the large number of depth layers.
AB - Dark matter simulations, performed using N-body methods with a finite set of tracer particles to discretize the initially uniform distribution of mass, are an invaluable method for exploring the formation of the universe. Definining a tetrahedral mesh in phase space-with the tracer particles at initialization serving as vertices-yields a more accurate density field. At later timesteps, the mesh self-intersects to an enormous degree, making pre-sorting impossible. Kaehler et al [2012] visualize the mesh using cell projection, but their method requires order-independent compositing, which limits its flexibility. Our work renders the mesh using state of the art order-independent transparency (OIT) techniques to composite fragments in correct depth order. This also allows us to render variables other than density, such as velocity. We implement a number of OIT optimizations to handle the high depth complexity (on the order of 107 depth layers for 2×109 particles) of the data. Our performance measurements show near-interactive framerates for our hybrid renderer despite the large number of depth layers.
KW - GPU acceleration
KW - Intersecting mesh
KW - Order-independent transparency
KW - Scientific visualization
KW - Volume rendering
UR - http://www.scopus.com/inward/record.url?scp=85006892859&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85006892859&partnerID=8YFLogxK
U2 - 10.1145/3002151.3002163
DO - 10.1145/3002151.3002163
M3 - Conference contribution
BT - SA 2016 - SIGGRAPH ASIA 2016 Symposium on Visualization
PB - Association for Computing Machinery, Inc
ER -