Proposals:GPU Integration: Difference between revisions
Danmueller (talk | contribs) m (Added Hessian-based vessel enhancement to wish list) |
No edit summary |
||
| Line 1: | Line 1: | ||
__TOC__ | __TOC__ | ||
= The Opportunity | = The Opportunity | ||
GPUs offer the opportunity for accelerating the execution of certain ITK filters. | GPUs offer the opportunity for accelerating the execution of certain ITK filters. | ||
= Proposal= | = Proposal= | ||
In collaboration with [http://www.interactivesupercomputing.com/ Interactive Supercomputing] we are looking into the possibility of adding GPU acceleration to particular algorithms in ITK. | |||
In collaboration with | |||
== The Methodology == | == The Methodology == | ||
The methodology suggested so far, is to take advantage of the Factory mechanism built-in ITK to create specialized versions of specific image filter. These | The methodology suggested so far, is to take advantage of the Factory mechanism built-in ITK to create specialized versions of specific image filter. These | ||
specialized versions will take advantage of a GPU-based implementation. | specialized versions will take advantage of a GPU-based implementation. | ||
| Line 18: | Line 15: | ||
== Resources == | == Resources == | ||
* [http://www.nvidia.com/object/cuda_home.html NVIDA CUDA Home] | * [http://www.nvidia.com/object/cuda_home.html NVIDA CUDA Home] | ||
* [http://developer.download.nvidia.com/compute/cuda/Photoshop/CUDAFilters4.pdf Building Photoshop Plugins in CUDA] | |||
* [http://sourceforge.net/projects/amdctm ATI CTM Sourceforge project] | * [http://sourceforge.net/projects/amdctm ATI CTM Sourceforge project] | ||
* [http://ati.amd.com/companyinfo/researcher/documents/ATI_CTM_Guide.pdf ATI CTM Manual] | * [http://ati.amd.com/companyinfo/researcher/documents/ATI_CTM_Guide.pdf ATI CTM Manual] | ||
* [http://techresearch.intel.com/articles/Tera-Scale/1514.htm Intel Ct Home] | |||
= Wishlist = | = Wishlist = | ||
We are looking for your feedback to define the list of algorithms that are good candidates for GPU acceleration. | We are looking for your feedback to define the list of algorithms that are good candidates for GPU acceleration. | ||
| Line 30: | Line 30: | ||
* Level Sets | * Level Sets | ||
* Hessian-based vessel enhancement (ie. Hessian Recursive Gaussian and Hessian 3D To Vesselness Measure) | * Hessian-based vessel enhancement (ie. Hessian Recursive Gaussian and Hessian 3D To Vesselness Measure) | ||
* Image smoothing | |||
Revision as of 17:21, 14 November 2008
= The Opportunity GPUs offer the opportunity for accelerating the execution of certain ITK filters.
Proposal
In collaboration with Interactive Supercomputing we are looking into the possibility of adding GPU acceleration to particular algorithms in ITK.
The Methodology
The methodology suggested so far, is to take advantage of the Factory mechanism built-in ITK to create specialized versions of specific image filter. These specialized versions will take advantage of a GPU-based implementation.
Since the acceleration requires to hand-craft the code, this will only be implemented for a few selected algorithms that are considered to be critical for the benefit of the ITK community.
Resources
Wishlist
We are looking for your feedback to define the list of algorithms that are good candidates for GPU acceleration.
Please add here the types of applications that you would consider to be good candidates for creating GPU-based specialized implementations.
- Deformable registration
- Level Sets
- Hessian-based vessel enhancement (ie. Hessian Recursive Gaussian and Hessian 3D To Vesselness Measure)
- Image smoothing
