Abstract
|
|
---|---|
Current GPU architectures provide impressive processing rates in graphical applications because of their specialized graphics pipeline. However, little attention has been paid to the analysis and study of different hardware architectures to implement specific pipeline stages. In this work we have identified one of the key stages in the graphics pipeline, the triangle traversal procedure, and we have imple- mented three different algorithms in hardware: bounding-box, zig-zag and Hilbert curve-based. The experimental results show that important area-performance trade-offs can be met when implementing key image processing algorithms in hardware. | |
International
|
Si |
Congress
|
25th Conference on Design of Circuits and Integrated Systems (DCIS'10) |
|
960 |
Place
|
Lanzarote, Spain |
Reviewers
|
Si |
ISBN/ISSN
|
9788469373934 |
|
|
Start Date
|
17/11/2010 |
End Date
|
19/11/2010 |
From page
|
54 |
To page
|
59 |
|
25th Conference on Design of Circuits and Integrated Systems Proceedings |