Mostra el registre d'ítem simple
Discriminative learning of deep convolutional feature point descriptors
dc.contributor.author | Simó Serra, Edgar |
dc.contributor.author | Trulls Fortuny, Eduard |
dc.contributor.author | Ferraz, Luis |
dc.contributor.author | Kokkinos, Iasonas |
dc.contributor.author | Fua, Pascal |
dc.contributor.author | Moreno-Noguer, Francesc |
dc.contributor.other | Institut de Robòtica i Informàtica Industrial |
dc.date.accessioned | 2016-03-11T19:09:33Z |
dc.date.available | 2016-03-11T19:09:33Z |
dc.date.issued | 2015 |
dc.identifier.citation | Simo, E., Trulls, E., Ferraz, L., Kokkinos, I., Fua, P., Moreno-Noguer, F. Discriminative learning of deep convolutional feature point descriptors. A: International Conference on Computer Vision. "Computer Vision (ICCV), 2015 IEEE International Conference on". Santiago de Chile: Institute of Electrical and Electronics Engineers (IEEE), 2015, p. 118-126. |
dc.identifier.uri | http://hdl.handle.net/2117/84259 |
dc.description.abstract | Deep learning has revolutionalized image-level tasks such as classification, but patch-level tasks, such as correspondence, still rely on hand-crafted features, e.g. SIFT. In this paper we use Convolutional Neural Networks (CNNs) to learn discriminant patch representations and in particular train a Siamese network with pairs of (non)corresponding patches. We deal with the large number of potential pairs with the combination of a stochastic sampling of the training set and an aggressive mining strategy biased towards patches that are hard to classify. By using the L2 distance during both training and test- ing we develop 128-D descriptors whose euclidean distances reflect patch similarity, and which can be used as a drop-in replacement for any task involving SIFT. We demonstrate consistent performance gains over the state of the art, and generalize well against scaling and rotation, perspective transformation, non-rigid deformation, and illumination changes. Our descriptors are efficient to compute and amenable to modern GPUs, and are publicly available. |
dc.format.extent | 9 p. |
dc.language.iso | eng |
dc.publisher | Institute of Electrical and Electronics Engineers (IEEE) |
dc.rights.uri | http://creativecommons.org/licenses/by-nc-nd/3.0/es/ |
dc.subject | Àrees temàtiques de la UPC::Informàtica::Automàtica i control |
dc.subject.other | computer vision |
dc.subject.other | feature extraction |
dc.subject.other | deep learning |
dc.title | Discriminative learning of deep convolutional feature point descriptors |
dc.type | Conference report |
dc.contributor.group | Universitat Politècnica de Catalunya. ROBiri - Grup de Robòtica de l'IRI |
dc.identifier.doi | 10.1109/ICCV.2015.22 |
dc.description.peerreviewed | Peer Reviewed |
dc.subject.inspec | Classificació INSPEC::Pattern recognition::Computer vision |
dc.relation.publisherversion | http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7410379 |
dc.rights.access | Open Access |
local.identifier.drac | 17556324 |
dc.description.version | Postprint (author's final draft) |
dc.relation.projectid | info:eu-repo/grantAgreement/EC/FP7/600796/EU/Intelligent Active MObility Aid RoBOT integrating Multimodal Communication/MOBOT |
dc.relation.projectid | info:eu-repo/grantAgreement/EC/H2020/644271/EU/AErial RObotic system integrating multiple ARMS and advanced manipulation capabilities for inspection and maintenance/AEROARMS |
dc.relation.projectid | info:eu-repo/grantAgreement/EC/H2020/643666/EU/ICT-Supported Bath Robots/I-SUPPORT |
local.citation.author | Simo, E.; Trulls, E.; Ferraz, L.; Kokkinos, I.; Fua, P.; Moreno-Noguer, F. |
local.citation.contributor | International Conference on Computer Vision |
local.citation.pubplace | Santiago de Chile |
local.citation.publicationName | Computer Vision (ICCV), 2015 IEEE International Conference on |
local.citation.startingPage | 118 |
local.citation.endingPage | 126 |