Speeded up robust features (SURF) can detect/describe scale- and rotation-invariant features at high speed by relying on integral images for image convolutions. However, the time taken for matching SURF descriptors is still long, and this has been an obstacle for use in real-time applications. In addition, the matching time further increases in proportion to the number of features and the dimensionality of the descriptor. Therefore, we propose a fast matching method that rearranges the elements of SURF descriptors based on their entropies, divides SURF descriptors into sub-descriptors, and sequentially and analytically matches them to each other. Our results show that the matching time could be reduced by about 75% at the expense of a small drop in accuracy.
Hanhoon PARK
Pukyong National University
Kwang-Seok MOON
Pukyong National University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Hanhoon PARK, Kwang-Seok MOON, "Fast Feature Matching by Coarse-to-Fine Comparison of Rearranged SURF Descriptors" in IEICE TRANSACTIONS on Information,
vol. E98-D, no. 1, pp. 210-213, January 2015, doi: 10.1587/transinf.2014EDL8149.
Abstract: Speeded up robust features (SURF) can detect/describe scale- and rotation-invariant features at high speed by relying on integral images for image convolutions. However, the time taken for matching SURF descriptors is still long, and this has been an obstacle for use in real-time applications. In addition, the matching time further increases in proportion to the number of features and the dimensionality of the descriptor. Therefore, we propose a fast matching method that rearranges the elements of SURF descriptors based on their entropies, divides SURF descriptors into sub-descriptors, and sequentially and analytically matches them to each other. Our results show that the matching time could be reduced by about 75% at the expense of a small drop in accuracy.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2014EDL8149/_p
Copy
@ARTICLE{e98-d_1_210,
author={Hanhoon PARK, Kwang-Seok MOON, },
journal={IEICE TRANSACTIONS on Information},
title={Fast Feature Matching by Coarse-to-Fine Comparison of Rearranged SURF Descriptors},
year={2015},
volume={E98-D},
number={1},
pages={210-213},
abstract={Speeded up robust features (SURF) can detect/describe scale- and rotation-invariant features at high speed by relying on integral images for image convolutions. However, the time taken for matching SURF descriptors is still long, and this has been an obstacle for use in real-time applications. In addition, the matching time further increases in proportion to the number of features and the dimensionality of the descriptor. Therefore, we propose a fast matching method that rearranges the elements of SURF descriptors based on their entropies, divides SURF descriptors into sub-descriptors, and sequentially and analytically matches them to each other. Our results show that the matching time could be reduced by about 75% at the expense of a small drop in accuracy.},
keywords={},
doi={10.1587/transinf.2014EDL8149},
ISSN={1745-1361},
month={January},}
Copy
TY - JOUR
TI - Fast Feature Matching by Coarse-to-Fine Comparison of Rearranged SURF Descriptors
T2 - IEICE TRANSACTIONS on Information
SP - 210
EP - 213
AU - Hanhoon PARK
AU - Kwang-Seok MOON
PY - 2015
DO - 10.1587/transinf.2014EDL8149
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E98-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 2015
AB - Speeded up robust features (SURF) can detect/describe scale- and rotation-invariant features at high speed by relying on integral images for image convolutions. However, the time taken for matching SURF descriptors is still long, and this has been an obstacle for use in real-time applications. In addition, the matching time further increases in proportion to the number of features and the dimensionality of the descriptor. Therefore, we propose a fast matching method that rearranges the elements of SURF descriptors based on their entropies, divides SURF descriptors into sub-descriptors, and sequentially and analytically matches them to each other. Our results show that the matching time could be reduced by about 75% at the expense of a small drop in accuracy.
ER -