dc.contributor.author |
Aguilera Carrasco, Cristhian Alejandro |
dc.contributor.author |
Toledo Morales, Ricardo |
dc.contributor.author |
Sappa, Angel Domingo |
dc.contributor.author |
Aguilera, Cristhian |
dc.date |
2017 |
dc.identifier |
https://ddd.uab.cat/record/181633 |
dc.identifier |
urn:10.3390/s17040873 |
dc.identifier |
urn:oai:ddd.uab.cat:181633 |
dc.identifier |
urn:pmid:28420142 |
dc.identifier |
urn:articleid:14248220v17n4p873 |
dc.identifier |
urn:scopus_id:85018505364 |
dc.identifier |
urn:wos_id:000400822900214 |
dc.identifier |
urn:oai:egreta.uab.cat:publications/2e4a5e17-ddf9-4e0e-8a2b-fc37af02eef2 |
dc.identifier |
urn:pmc-uid:5424750 |
dc.identifier |
urn:pmcid:PMC5424750 |
dc.identifier |
urn:oai:pubmedcentral.nih.gov:5424750 |
dc.format |
application/pdf |
dc.language |
eng |
dc.publisher |
|
dc.relation |
Ministerio de Economía y Competitividad TIN2014-56919-C3-2-R |
dc.relation |
Sensors (Basel, Switzerland) ; Vol. 17 Núm. 4 (April 2017), art. 873 |
dc.rights |
open access |
dc.rights |
Aquest document està subjecte a una llicència d'ús Creative Commons. Es permet la reproducció total o parcial, la distribució, la comunicació pública de l'obra i la creació d'obres derivades, fins i tot amb finalitats comercials, sempre i quan es reconegui l'autoria de l'obra original. |
dc.rights |
https://creativecommons.org/licenses/by/4.0/ |
dc.subject |
Cross-spectral |
dc.subject |
Descriptor |
dc.subject |
Infrared |
dc.subject |
CNN |
dc.title |
Cross-spectral local descriptors via quadruplet network |
dc.type |
Article |
dc.description.abstract |
This paper presents a novel CNN-based architecture, referred to as Q-Net, to learn local feature descriptors that are useful for matching image patches from two different spectral bands. Given correctly matched and non-matching cross-spectral image pairs, a quadruplet network is trained to map input image patches to a common Euclidean space, regardless of the input spectral band. Our approach is inspired by the recent success of triplet networks in the visible spectrum, but adapted for cross-spectral scenarios, where, for each matching pair, there are always two possible non-matching patches: one for each spectrum. Experimental evaluations on a public cross-spectral VIS-NIR dataset shows that the proposed approach improves the state-of-the-art. Moreover, the proposed technique can also be used in mono-spectral settings, obtaining a similar performance to triplet network descriptors, but requiring less training data. |