liu.seSearch for publications in DiVA
Change search
ReferencesLink to record
Permanent link

Direct link
Optical Flow Computation on CUDA
Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, The Institute of Technology.
2009 (English)In: SSBA, 2009, 81-84 p.Conference paper (Other academic)
Abstract [en]

There has been a rapid progress of the graphics processor the last years, much because of the demands from computer games on speed and realistic rendering. Because of the graphics processor’s special architecture it is much faster at solving parallel problems than the normal processor. Due to its increasing generality it is possible to use it for other tasks than it was originally designed for.

Even though graphics processors have been programmable for some time, it has been quite difficult to learn how to use them. CUDA (Compute Unified Device Architecture) enables the programmer to use C-code, with a few extensions, to program NVIDIA’s graphics processor and completely skip the traditional programming models. This paper investigates if the graphics processor can be used for calculations without knowledge of how the hardware mechanisms work. An image processing algorithm calculating the optical flow has been implemented. The result shows that it is rather easy to implement programs using CUDA, but some knowledge of how the graphics processor works is required to achieve high performance.

Place, publisher, year, edition, pages
2009. 81-84 p.
National Category
Engineering and Technology
URN: urn:nbn:se:liu:diva-50617OAI: diva2:271770
Available from: 2009-10-13 Created: 2009-10-13

Open Access in DiVA

No full text

Search in DiVA

By author/editor
Ringaby, Erik
By organisation
Computer VisionThe Institute of Technology
Engineering and Technology

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 237 hits
ReferencesLink to record
Permanent link

Direct link