FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3105159632> ?p ?o ?g. }

• W3105159632 endingPage "3153" @default.
• W3105159632 startingPage "3140" @default.
• W3105159632 abstract "Currently, attention mechanism receives enormous interest and has been extensively employed in the fields of Person Re-Identification (RE-ID), as it gains superior performance in learning discriminative feature representations. However, most off-the-shelf attention methods are still vulnerable to cross-view inconsistency problem. Besides, they merely exploit imprecise channel attention information and coarse-grained spatial attention of homogeneous scales, being insufficient to capture subtle differences among highly-similar individuals. To this end, we propose a novel Attention-Aligned Network (AANet) to address the aforementioned problems, in which a novel Omnibearing Foreground-aware Attention (OFA) module, Attention Alignment Mechanism (AAM) and an improved triplet loss with hard mining are proposed to learn foreground attentive features for RE-ID. Specifically, AANet firstly leverages OFA module to exploit heterogeneous-scale spatial attention and foreground-aware channel attention information. Then AANet further reduces the impact of background clutter and learns camera-invariant and background-invariant representations by virtue of AAM. Last but not least, an improved triplet loss with hard mining is also introduced to enhance the feature learning capability, which can jointly minimize the intra-class distance and maximize the inter-class distance in each triplet unit. Extensive experiments are carried out to demonstrate that the proposed method outperforms most current methods on three main RE-ID benchmarks." @default.
• W3105159632 created "2020-11-23" @default.
• W3105159632 creator A5006676654 @default.
• W3105159632 creator A5029203943 @default.
• W3105159632 creator A5042991224 @default.
• W3105159632 date "2021-08-01" @default.
• W3105159632 modified "2023-10-14" @default.
• W3105159632 title "Attention-Aligned Network for Person Re-Identification" @default.
• W3105159632 cites W1971955426 @default.
• W3105159632 cites W1982925187 @default.
• W3105159632 cites W2064675550 @default.
• W3105159632 cites W2183341477 @default.
• W3105159632 cites W2204750386 @default.
• W3105159632 cites W2295107390 @default.
• W3105159632 cites W2511791013 @default.
• W3105159632 cites W2550553598 @default.
• W3105159632 cites W2551012455 @default.
• W3105159632 cites W2584637367 @default.
• W3105159632 cites W2585635281 @default.
• W3105159632 cites W2606377603 @default.
• W3105159632 cites W2752782242 @default.
• W3105159632 cites W2765268259 @default.
• W3105159632 cites W2765995039 @default.
• W3105159632 cites W2768610172 @default.
• W3105159632 cites W2798794112 @default.
• W3105159632 cites W2798874329 @default.
• W3105159632 cites W2799047769 @default.
• W3105159632 cites W2804236575 @default.
• W3105159632 cites W2896888563 @default.
• W3105159632 cites W2962691289 @default.
• W3105159632 cites W2962706983 @default.
• W3105159632 cites W2962926870 @default.
• W3105159632 cites W2962928529 @default.
• W3105159632 cites W2963446712 @default.
• W3105159632 cites W2963736028 @default.
• W3105159632 cites W2963805953 @default.
• W3105159632 cites W2963811535 @default.
• W3105159632 cites W2963842104 @default.
• W3105159632 cites W2963910742 @default.
• W3105159632 cites W2963988212 @default.
• W3105159632 cites W2964140013 @default.
• W3105159632 cites W2964344300 @default.
• W3105159632 cites W2980337913 @default.
• W3105159632 cites W2981420411 @default.
• W3105159632 cites W2998493658 @default.
• W3105159632 cites W3009734829 @default.
• W3105159632 cites W3034580371 @default.
• W3105159632 cites W3035186652 @default.
• W3105159632 cites W3098420177 @default.
• W3105159632 cites W3124951096 @default.
• W3105159632 doi "https://doi.org/10.1109/tcsvt.2020.3037179" @default.
• W3105159632 hasPublicationYear "2021" @default.
• W3105159632 type Work @default.
• W3105159632 sameAs 3105159632 @default.
• W3105159632 citedByCount "20" @default.
• W3105159632 countsByYear W31051596322021 @default.
• W3105159632 countsByYear W31051596322022 @default.
• W3105159632 countsByYear W31051596322023 @default.
• W3105159632 crossrefType "journal-article" @default.
• W3105159632 hasAuthorship W3105159632A5006676654 @default.
• W3105159632 hasAuthorship W3105159632A5029203943 @default.
• W3105159632 hasAuthorship W3105159632A5042991224 @default.
• W3105159632 hasConcept C116834253 @default.
• W3105159632 hasConcept C119857082 @default.
• W3105159632 hasConcept C121332964 @default.
• W3105159632 hasConcept C132094186 @default.
• W3105159632 hasConcept C138885662 @default.
• W3105159632 hasConcept C153180895 @default.
• W3105159632 hasConcept C154945302 @default.
• W3105159632 hasConcept C165696696 @default.
• W3105159632 hasConcept C190470478 @default.
• W3105159632 hasConcept C2776401178 @default.
• W3105159632 hasConcept C37914503 @default.
• W3105159632 hasConcept C38652104 @default.
• W3105159632 hasConcept C41008148 @default.
• W3105159632 hasConcept C41895202 @default.
• W3105159632 hasConcept C48044578 @default.
• W3105159632 hasConcept C554190296 @default.
• W3105159632 hasConcept C59404180 @default.
• W3105159632 hasConcept C59822182 @default.
• W3105159632 hasConcept C76155785 @default.
• W3105159632 hasConcept C77088390 @default.
• W3105159632 hasConcept C86803240 @default.
• W3105159632 hasConcept C97931131 @default.
• W3105159632 hasConceptScore W3105159632C116834253 @default.
• W3105159632 hasConceptScore W3105159632C119857082 @default.
• W3105159632 hasConceptScore W3105159632C121332964 @default.
• W3105159632 hasConceptScore W3105159632C132094186 @default.
• W3105159632 hasConceptScore W3105159632C138885662 @default.
• W3105159632 hasConceptScore W3105159632C153180895 @default.
• W3105159632 hasConceptScore W3105159632C154945302 @default.
• W3105159632 hasConceptScore W3105159632C165696696 @default.
• W3105159632 hasConceptScore W3105159632C190470478 @default.
• W3105159632 hasConceptScore W3105159632C2776401178 @default.
• W3105159632 hasConceptScore W3105159632C37914503 @default.
• W3105159632 hasConceptScore W3105159632C38652104 @default.
• W3105159632 hasConceptScore W3105159632C41008148 @default.
• W3105159632 hasConceptScore W3105159632C41895202 @default.

• next