Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Feature Detection and Description using a Harris-Hessian/FREAK Combination on an Embedded GPU

Danielsson, Max ; Sievert, Thomas ; Grahn, Håkan and Ramusson, Jim (2016) 5th International Conference on Pattern Recognition Applications and Methods (IPCRAM 2016) p.517-525
Abstract
GPUs in embedded platforms are reaching performance levels comparable to desktop hardware, thus it becomes interesting to apply Computer Vision techniques. We propose, implement, and evaluate a novel feature detector and descriptor combination, i.e., we combine the Harris-Hessian detector with the FREAK binary descriptor. The implementation is done in OpenCL, and we evaluate the execution time and classification performance. We compare our approach with two other methods, FAST/BRISK and ORB. Performance data is presented for the mobile device Xperia Z3 and the desktop Nvidia GTX 660. Our results indicate that the execution times on the Xperia Z3 are insufficient for real-time applications while desktop execution shows future potential.... (More)
GPUs in embedded platforms are reaching performance levels comparable to desktop hardware, thus it becomes interesting to apply Computer Vision techniques. We propose, implement, and evaluate a novel feature detector and descriptor combination, i.e., we combine the Harris-Hessian detector with the FREAK binary descriptor. The implementation is done in OpenCL, and we evaluate the execution time and classification performance. We compare our approach with two other methods, FAST/BRISK and ORB. Performance data is presented for the mobile device Xperia Z3 and the desktop Nvidia GTX 660. Our results indicate that the execution times on the Xperia Z3 are insufficient for real-time applications while desktop execution shows future potential. Classification performance of Harris-Hessian/FREAK indicates that the solution is sensitive to rotation, but superior in scale variant images. (Less)
Please use this url to cite or link to this publication:
author
; ; and
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
host publication
Proceedings of the 5th International Conference on Pattern Recognition Applications and Methods
pages
517 - 525
publisher
SciTePress
conference name
5th International Conference on Pattern Recognition Applications and Methods (IPCRAM 2016)
conference location
Rome, Italy
conference dates
2016-02-24 - 2016-02-26
external identifiers
  • scopus:84969931851
ISBN
978-989-758-173-1
DOI
10.5220/0005662005170525
project
Embedded Applications Software Engineering
language
English
LU publication?
no
id
eef9fdb0-fbe5-45cb-9300-ef7405d18f84
date added to LUP
2018-09-25 11:47:12
date last changed
2024-01-15 01:56:45
@inproceedings{eef9fdb0-fbe5-45cb-9300-ef7405d18f84,
  abstract     = {{GPUs in embedded platforms are reaching performance levels comparable to desktop hardware, thus it becomes interesting to apply Computer Vision techniques. We propose, implement, and evaluate a novel feature detector and descriptor combination, i.e., we combine the Harris-Hessian detector with the FREAK binary descriptor. The implementation is done in OpenCL, and we evaluate the execution time and classification performance. We compare our approach with two other methods, FAST/BRISK and ORB. Performance data is presented for the mobile device Xperia Z3 and the desktop Nvidia GTX 660. Our results indicate that the execution times on the Xperia Z3 are insufficient for real-time applications while desktop execution shows future potential. Classification performance of Harris-Hessian/FREAK indicates that the solution is sensitive to rotation, but superior in scale variant images.}},
  author       = {{Danielsson, Max and Sievert, Thomas and Grahn, Håkan and Ramusson, Jim}},
  booktitle    = {{Proceedings of the 5th International Conference on Pattern Recognition Applications and Methods}},
  isbn         = {{978-989-758-173-1}},
  language     = {{eng}},
  pages        = {{517--525}},
  publisher    = {{SciTePress}},
  title        = {{Feature Detection and Description using a Harris-Hessian/FREAK Combination on an Embedded GPU}},
  url          = {{http://dx.doi.org/10.5220/0005662005170525}},
  doi          = {{10.5220/0005662005170525}},
  year         = {{2016}},
}