Picking the best DAISY

S Winder, Gang Hua, Matthew A Brown

Research output: Chapter in Book/Report/Conference proceedingConference contribution

212 Citations (Scopus)
87 Downloads (Pure)

Abstract

Local image descriptors that are highly discriminative, computational efficient, and with low storage footprint have long been a dream goal of computer vision research. In this paper, we focus on learning such descriptors, which make use of the DAISY configuration and are simple to compute both sparsely and densely. We develop a new training set of match/non-match image patches which improves on previous work. We test a wide variety of gradient and steerable filter based configurations and optimize over all parameters to obtain low matching errors for the descriptors. We further explore robust normalization, dimension reduction and dynamic range reduction to increase the discriminative power and yet reduce the storage requirement of the learned descriptors. All these enable us to obtain highly efficient local descriptors: e.g, 13.2% error at 13 bytes storage per descriptor, compared with 26.1% error at 128 bytes for SIFT.
Original languageEnglish
Title of host publication2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2009)
PublisherIEEE
Pages178-185
Number of pages8
ISBN (Print)9781424439928, 9781424439911
DOIs
Publication statusPublished - 13 Aug 2009
EventCVPR 2009: IEEE Conference on Computer Vision and Pattern Recognition, 2009 - Miami
Duration: 20 Jun 200925 Jun 2009

Conference

ConferenceCVPR 2009: IEEE Conference on Computer Vision and Pattern Recognition, 2009
CityMiami
Period20/06/0925/06/09

Fingerprint

Computer vision

Cite this

Winder, S., Hua, G., & Brown, M. A. (2009). Picking the best DAISY. In 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2009) (pp. 178-185). IEEE. https://doi.org/10.1109/CVPR.2009.5206839

Picking the best DAISY. / Winder, S; Hua, Gang; Brown, Matthew A.

2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2009). IEEE, 2009. p. 178-185.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Winder, S, Hua, G & Brown, MA 2009, Picking the best DAISY. in 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2009). IEEE, pp. 178-185, CVPR 2009: IEEE Conference on Computer Vision and Pattern Recognition, 2009, Miami, 20/06/09. https://doi.org/10.1109/CVPR.2009.5206839
Winder S, Hua G, Brown MA. Picking the best DAISY. In 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2009). IEEE. 2009. p. 178-185 https://doi.org/10.1109/CVPR.2009.5206839
Winder, S ; Hua, Gang ; Brown, Matthew A. / Picking the best DAISY. 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2009). IEEE, 2009. pp. 178-185
@inproceedings{6fe888a4c97945a8971d6a2479927dc0,
title = "Picking the best DAISY",
abstract = "Local image descriptors that are highly discriminative, computational efficient, and with low storage footprint have long been a dream goal of computer vision research. In this paper, we focus on learning such descriptors, which make use of the DAISY configuration and are simple to compute both sparsely and densely. We develop a new training set of match/non-match image patches which improves on previous work. We test a wide variety of gradient and steerable filter based configurations and optimize over all parameters to obtain low matching errors for the descriptors. We further explore robust normalization, dimension reduction and dynamic range reduction to increase the discriminative power and yet reduce the storage requirement of the learned descriptors. All these enable us to obtain highly efficient local descriptors: e.g, 13.2{\%} error at 13 bytes storage per descriptor, compared with 26.1{\%} error at 128 bytes for SIFT.",
author = "S Winder and Gang Hua and Brown, {Matthew A}",
year = "2009",
month = "8",
day = "13",
doi = "10.1109/CVPR.2009.5206839",
language = "English",
isbn = "9781424439928",
pages = "178--185",
booktitle = "2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2009)",
publisher = "IEEE",
address = "USA United States",

}

TY - GEN

T1 - Picking the best DAISY

AU - Winder, S

AU - Hua, Gang

AU - Brown, Matthew A

PY - 2009/8/13

Y1 - 2009/8/13

N2 - Local image descriptors that are highly discriminative, computational efficient, and with low storage footprint have long been a dream goal of computer vision research. In this paper, we focus on learning such descriptors, which make use of the DAISY configuration and are simple to compute both sparsely and densely. We develop a new training set of match/non-match image patches which improves on previous work. We test a wide variety of gradient and steerable filter based configurations and optimize over all parameters to obtain low matching errors for the descriptors. We further explore robust normalization, dimension reduction and dynamic range reduction to increase the discriminative power and yet reduce the storage requirement of the learned descriptors. All these enable us to obtain highly efficient local descriptors: e.g, 13.2% error at 13 bytes storage per descriptor, compared with 26.1% error at 128 bytes for SIFT.

AB - Local image descriptors that are highly discriminative, computational efficient, and with low storage footprint have long been a dream goal of computer vision research. In this paper, we focus on learning such descriptors, which make use of the DAISY configuration and are simple to compute both sparsely and densely. We develop a new training set of match/non-match image patches which improves on previous work. We test a wide variety of gradient and steerable filter based configurations and optimize over all parameters to obtain low matching errors for the descriptors. We further explore robust normalization, dimension reduction and dynamic range reduction to increase the discriminative power and yet reduce the storage requirement of the learned descriptors. All these enable us to obtain highly efficient local descriptors: e.g, 13.2% error at 13 bytes storage per descriptor, compared with 26.1% error at 128 bytes for SIFT.

UR - http://www.scopus.com/inward/record.url?scp=70450208928&partnerID=8YFLogxK

UR - http://dx.doi.org/10.1109/CVPR.2009.5206839

U2 - 10.1109/CVPR.2009.5206839

DO - 10.1109/CVPR.2009.5206839

M3 - Conference contribution

SN - 9781424439928

SN - 9781424439911

SP - 178

EP - 185

BT - 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2009)

PB - IEEE

ER -