Efficient Integral Image Computation on the GPU
We present an integral image algorithm that can run in real-time on a Graphics Processing Unit (GPU). Our system exploits the parallelisms in computation via the NIVIDA CUDA programming model, which is a software platform for solving non-graphics problems in a massively parallel high-performance fas...
| Principais autores: | , , |
|---|---|
| Outros Autores: | |
| Formato: | Artigo |
| Idioma: | en_US |
| Publicado em: |
Institute of Electrical and Electronics Engineers (IEEE)
2012
|
| Acesso em linha: | http://hdl.handle.net/1721.1/71883 https://orcid.org/0000-0003-3434-391X https://orcid.org/0000-0002-6657-5646 |
| _version_ | 1826216868678991872 |
|---|---|
| author | Bilgic, Berkin Horn, Berthold Klaus Paul Masaki, Ichiro |
| author2 | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science |
| author_facet | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Bilgic, Berkin Horn, Berthold Klaus Paul Masaki, Ichiro |
| author_sort | Bilgic, Berkin |
| collection | MIT |
| description | We present an integral image algorithm that can run in real-time on a Graphics Processing Unit (GPU). Our system exploits the parallelisms in computation via the NIVIDA CUDA programming model, which is a software platform for solving non-graphics problems in a massively parallel high-performance fashion. This implementation makes use of the work-efficient scan algorithm that is explicated in. Treating the rows and the columns of the target image as independent input arrays for the scan algorithm, our method manages to expose a second level of parallelism in the problem. We compare the performance of the parallel approach running on the GPU with the sequential CPU implementation across a range of image sizes and report a speed up by a factor of 8 for a 4 megapixel input. We further investigate the impact of using packed vector type data on the performance, as well as the effect of double precision arithmetic on the GPU. |
| first_indexed | 2024-09-23T16:54:23Z |
| format | Article |
| id | mit-1721.1/71883 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T16:54:23Z |
| publishDate | 2012 |
| publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| record_format | dspace |
| spelling | mit-1721.1/718832022-09-29T22:19:04Z Efficient Integral Image Computation on the GPU Bilgic, Berkin Horn, Berthold Klaus Paul Masaki, Ichiro Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Horn, Berthold K. P. Bilgic, Berkin Horn, Berthold Klaus Paul Masaki, Ichiro We present an integral image algorithm that can run in real-time on a Graphics Processing Unit (GPU). Our system exploits the parallelisms in computation via the NIVIDA CUDA programming model, which is a software platform for solving non-graphics problems in a massively parallel high-performance fashion. This implementation makes use of the work-efficient scan algorithm that is explicated in. Treating the rows and the columns of the target image as independent input arrays for the scan algorithm, our method manages to expose a second level of parallelism in the problem. We compare the performance of the parallel approach running on the GPU with the sequential CPU implementation across a range of image sizes and report a speed up by a factor of 8 for a 4 megapixel input. We further investigate the impact of using packed vector type data on the performance, as well as the effect of double precision arithmetic on the GPU. 2012-07-30T12:51:55Z 2012-07-30T12:51:55Z 2010-06 2010-05 Article http://purl.org/eprint/type/ConferencePaper 978-1-4244-7866-8 1931-0587 http://hdl.handle.net/1721.1/71883 Bilgic, B, B K P Horn, and I Masaki. “Efficient Integral Image Computation on the GPU.” IEEE, 2010. 528–533. © Copyright 2010 IEEE https://orcid.org/0000-0003-3434-391X https://orcid.org/0000-0002-6657-5646 en_US http://dx.doi.org/10.1109/IVS.2010.5548142 2010 IEEE Intelligent Vehicles Symposium (IV) Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. application/pdf Institute of Electrical and Electronics Engineers (IEEE) IEEE |
| spellingShingle | Bilgic, Berkin Horn, Berthold Klaus Paul Masaki, Ichiro Efficient Integral Image Computation on the GPU |
| title | Efficient Integral Image Computation on the GPU |
| title_full | Efficient Integral Image Computation on the GPU |
| title_fullStr | Efficient Integral Image Computation on the GPU |
| title_full_unstemmed | Efficient Integral Image Computation on the GPU |
| title_short | Efficient Integral Image Computation on the GPU |
| title_sort | efficient integral image computation on the gpu |
| url | http://hdl.handle.net/1721.1/71883 https://orcid.org/0000-0003-3434-391X https://orcid.org/0000-0002-6657-5646 |
| work_keys_str_mv | AT bilgicberkin efficientintegralimagecomputationonthegpu AT hornbertholdklauspaul efficientintegralimagecomputationonthegpu AT masakiichiro efficientintegralimagecomputationonthegpu |