Please use this identifier to cite or link to this item:
https://hdl.handle.net/10316/93988
Title: | Fast Scale-Invariant Feature Transform on GPU | Other Titles: | Fast Scale-Invariant Feature Transform on GPU | Authors: | Barreiros, João Carlos da Costa | Orientador: | Fernandes, Gabriel Falcão Paiva | Keywords: | Feature extraction; Scale-invariant feature transform; GPGPU; CUDA; Parallel Programming; Feature extraction; Scale-invariant feature transform; GPGPU; CUDA; Parallel Programming | Issue Date: | 17-Dec-2020 | metadata.degois.publication.title: | Fast Scale-Invariant Feature Transform on GPU | metadata.degois.publication.location: | DEEC | Abstract: | 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. 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. |
Description: | Dissertação de Mestrado Integrado em Engenharia Electrotécnica e de Computadores apresentada à Faculdade de Ciências e Tecnologia | URI: | https://hdl.handle.net/10316/93988 | Rights: | openAccess |
Appears in Collections: | UC - Dissertações de Mestrado |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
DissertacaoJoaoBarreiros_23 versão definitiva.pdf | 12.74 MB | Adobe PDF | View/Open |
Page view(s)
109
checked on Nov 6, 2024
Download(s)
213
checked on Nov 6, 2024
Google ScholarTM
Check
This item is licensed under a Creative Commons License