Fast Scale-Invariant Feature Transform on GPU
| Autor(a) principal: | |
|---|---|
| Data de Publicação: | 2020 |
| Tipo de documento: | Dissertação |
| Idioma: | eng |
| Título da fonte: | Repositórios Científicos de Acesso Aberto de Portugal (RCAAP) |
| Texto Completo: | https://hdl.handle.net/10316/93988 |
Resumo: | Dissertação de Mestrado Integrado em Engenharia Electrotécnica e de Computadores apresentada à Faculdade de Ciências e Tecnologia |
| id |
RCAP_628f022a49073d7106b1f1ea7018e7de |
|---|---|
| oai_identifier_str |
oai:estudogeral.uc.pt:10316/93988 |
| network_acronym_str |
RCAP |
| network_name_str |
Repositórios Científicos de Acesso Aberto de Portugal (RCAAP) |
| repository_id_str |
https://opendoar.ac.uk/repository/7160 |
| spelling |
Fast Scale-Invariant Feature Transform on GPUFast Scale-Invariant Feature Transform on GPUFeature extractionScale-invariant feature transformGPGPUCUDAParallel ProgrammingFeature extractionScale-invariant feature transformGPGPUCUDAParallel ProgrammingDissertação de Mestrado Integrado em Engenharia Electrotécnica e de Computadores apresentada à Faculdade de Ciências e TecnologiaFeature extraction of high-resolution images is a challenging procedure in low-power signal processing applications. This thesis describes how to optimize and efficiently parallelize the scale-invariant feature transform (SIFT) feature detection algorithm and maximize the use of bandwidth on the GPUsubsystem. Together with the minimization of data communications between host and device, the successful parallelization of all the main kernels used in SIFT allowed a global speedup in high-resolution images above 78x while being more than an order of magnitude energy efficient (FPS/W) than its serial counterpart. From the 3 GPUs tested, the low-power GPU has shown superior energy efficiency -- 44 FPS/W.Feature extraction of high-resolution images is a challenging procedure in low-power signal processing applications. This thesis describes how to optimize and efficiently parallelize the scale-invariant feature transform (SIFT) feature detection algorithm and maximize the use of bandwidth on the GPUsubsystem. Together with the minimization of data communications between host and device, the successful parallelization of all the main kernels used in SIFT allowed a global speedup in high-resolution images above 78x while being more than an order of magnitude energy efficient (FPS/W) than its serial counterpart. From the 3 GPUs tested, the low-power GPU has shown superior energy efficiency -- 44 FPS/W.2020-12-17info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesishttps://hdl.handle.net/10316/93988https://hdl.handle.net/10316/93988TID:202686574engBarreiros, João Carlos da Costainfo:eu-repo/semantics/openAccessreponame:Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)instname:FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologiainstacron:RCAAP2022-05-25T10:25:54Zoai:estudogeral.uc.pt:10316/93988Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireinfo@rcaap.ptopendoar:https://opendoar.ac.uk/repository/71602025-05-29T05:41:59.373664Repositórios Científicos de Acesso Aberto de Portugal (RCAAP) - FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologiafalse |
| dc.title.none.fl_str_mv |
Fast Scale-Invariant Feature Transform on GPU Fast Scale-Invariant Feature Transform on GPU |
| title |
Fast Scale-Invariant Feature Transform on GPU |
| spellingShingle |
Fast Scale-Invariant Feature Transform on GPU Barreiros, João Carlos da Costa Feature extraction Scale-invariant feature transform GPGPU CUDA Parallel Programming Feature extraction Scale-invariant feature transform GPGPU CUDA Parallel Programming |
| title_short |
Fast Scale-Invariant Feature Transform on GPU |
| title_full |
Fast Scale-Invariant Feature Transform on GPU |
| title_fullStr |
Fast Scale-Invariant Feature Transform on GPU |
| title_full_unstemmed |
Fast Scale-Invariant Feature Transform on GPU |
| title_sort |
Fast Scale-Invariant Feature Transform on GPU |
| author |
Barreiros, João Carlos da Costa |
| author_facet |
Barreiros, João Carlos da Costa |
| author_role |
author |
| dc.contributor.author.fl_str_mv |
Barreiros, João Carlos da Costa |
| dc.subject.por.fl_str_mv |
Feature extraction Scale-invariant feature transform GPGPU CUDA Parallel Programming Feature extraction Scale-invariant feature transform GPGPU CUDA Parallel Programming |
| topic |
Feature extraction Scale-invariant feature transform GPGPU CUDA Parallel Programming Feature extraction Scale-invariant feature transform GPGPU CUDA Parallel Programming |
| description |
Dissertação de Mestrado Integrado em Engenharia Electrotécnica e de Computadores apresentada à Faculdade de Ciências e Tecnologia |
| publishDate |
2020 |
| dc.date.none.fl_str_mv |
2020-12-17 |
| dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
| dc.type.driver.fl_str_mv |
info:eu-repo/semantics/masterThesis |
| format |
masterThesis |
| status_str |
publishedVersion |
| dc.identifier.uri.fl_str_mv |
https://hdl.handle.net/10316/93988 https://hdl.handle.net/10316/93988 TID:202686574 |
| url |
https://hdl.handle.net/10316/93988 |
| identifier_str_mv |
TID:202686574 |
| dc.language.iso.fl_str_mv |
eng |
| language |
eng |
| dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
| eu_rights_str_mv |
openAccess |
| dc.source.none.fl_str_mv |
reponame:Repositórios Científicos de Acesso Aberto de Portugal (RCAAP) instname:FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia instacron:RCAAP |
| instname_str |
FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia |
| instacron_str |
RCAAP |
| institution |
RCAAP |
| reponame_str |
Repositórios Científicos de Acesso Aberto de Portugal (RCAAP) |
| collection |
Repositórios Científicos de Acesso Aberto de Portugal (RCAAP) |
| repository.name.fl_str_mv |
Repositórios Científicos de Acesso Aberto de Portugal (RCAAP) - FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia |
| repository.mail.fl_str_mv |
info@rcaap.pt |
| _version_ |
1833602442081599488 |