Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313591085> ?p ?o ?g. }
- W4313591085 endingPage "559" @default.
- W4313591085 startingPage "545" @default.
- W4313591085 abstract "Visible-infrared person re-identification (VI-ReID) is a challenging and essential task, which aims to retrieve a set of person images over visible and infrared camera views. In order to mitigate the impact of large modality discrepancy existing in heterogeneous images, previous methods attempt to apply generative adversarial network (GAN) to generate the modality-consisitent data. However, due to severe color variations between the visible domain and infrared domain, the generated fake cross-modality samples often fail to possess good qualities to fill the modality gap between synthesized scenarios and target real ones, which leads to sub-optimal feature representations. In this work, we address cross-modality matching problem with Aligned Grayscale Modality (AGM), an unified dark-line spectrum that reformulates visible-infrared dual-mode learning as a gray-gray single-mode learning problem. Specifically, we generate the grasycale modality from the homogeneous visible images. Then, we train a style tranfer model to transfer infrared images into homogeneous grayscale images. In this way, the modality discrepancy is significantly reduced in the image space. In order to reduce the remaining appearance discrepancy, we further introduce a multi-granularity feature extraction network to conduct feature-level alignment. Rather than relying on the global information, we propose to exploit local (head-shoulder) features to assist person Re-ID, which complements each other to form a stronger feature descriptor. Comprehensive experiments implemented on the mainstream evaluation datasets include SYSU-MM01 and RegDB indicate that our method can significantly boost cross-modality retrieval performance against the state of the art methods." @default.
- W4313591085 created "2023-01-06" @default.
- W4313591085 creator A5009301758 @default.
- W4313591085 creator A5035484030 @default.
- W4313591085 creator A5076624593 @default.
- W4313591085 date "2023-05-01" @default.
- W4313591085 modified "2023-09-29" @default.
- W4313591085 title "Towards Homogeneous Modality Learning and Multi-Granularity Information Exploration for Visible-Infrared Person Re-Identification" @default.
- W4313591085 cites W2502225121 @default.
- W4313591085 cites W2596603442 @default.
- W4313591085 cites W2604463754 @default.
- W4313591085 cites W2777534232 @default.
- W4313591085 cites W2788846526 @default.
- W4313591085 cites W2798429327 @default.
- W4313591085 cites W2808134123 @default.
- W4313591085 cites W2808260522 @default.
- W4313591085 cites W2810342182 @default.
- W4313591085 cites W2810967842 @default.
- W4313591085 cites W2912406707 @default.
- W4313591085 cites W2954773727 @default.
- W4313591085 cites W2962793481 @default.
- W4313591085 cites W2963597983 @default.
- W4313591085 cites W2967515867 @default.
- W4313591085 cites W2970390221 @default.
- W4313591085 cites W2971537151 @default.
- W4313591085 cites W2981393440 @default.
- W4313591085 cites W2985033611 @default.
- W4313591085 cites W2992183111 @default.
- W4313591085 cites W2997877744 @default.
- W4313591085 cites W2998522188 @default.
- W4313591085 cites W2998792609 @default.
- W4313591085 cites W3026581226 @default.
- W4313591085 cites W3033235266 @default.
- W4313591085 cites W3034084848 @default.
- W4313591085 cites W3034519219 @default.
- W4313591085 cites W3034580371 @default.
- W4313591085 cites W3035673257 @default.
- W4313591085 cites W3100506510 @default.
- W4313591085 cites W3106715802 @default.
- W4313591085 cites W3107848599 @default.
- W4313591085 cites W3134669182 @default.
- W4313591085 cites W3137802886 @default.
- W4313591085 cites W3155080005 @default.
- W4313591085 cites W3168126494 @default.
- W4313591085 cites W3173203577 @default.
- W4313591085 cites W3173469406 @default.
- W4313591085 cites W3176633985 @default.
- W4313591085 cites W3192055375 @default.
- W4313591085 cites W3195006847 @default.
- W4313591085 cites W3198713693 @default.
- W4313591085 cites W3202750592 @default.
- W4313591085 cites W3202788649 @default.
- W4313591085 cites W3204075450 @default.
- W4313591085 cites W3207576380 @default.
- W4313591085 cites W3207905476 @default.
- W4313591085 cites W3210946531 @default.
- W4313591085 cites W4200635168 @default.
- W4313591085 cites W4210481275 @default.
- W4313591085 cites W4226294111 @default.
- W4313591085 cites W4308487110 @default.
- W4313591085 cites W4312802153 @default.
- W4313591085 cites W4312934679 @default.
- W4313591085 cites W4312936309 @default.
- W4313591085 doi "https://doi.org/10.1109/jstsp.2022.3233716" @default.
- W4313591085 hasPublicationYear "2023" @default.
- W4313591085 type Work @default.
- W4313591085 citedByCount "4" @default.
- W4313591085 countsByYear W43135910852023 @default.
- W4313591085 crossrefType "journal-article" @default.
- W4313591085 hasAuthorship W4313591085A5009301758 @default.
- W4313591085 hasAuthorship W4313591085A5035484030 @default.
- W4313591085 hasAuthorship W4313591085A5076624593 @default.
- W4313591085 hasBestOaLocation W43135910852 @default.
- W4313591085 hasConcept C111919701 @default.
- W4313591085 hasConcept C138885662 @default.
- W4313591085 hasConcept C150899416 @default.
- W4313591085 hasConcept C153180895 @default.
- W4313591085 hasConcept C154945302 @default.
- W4313591085 hasConcept C160633673 @default.
- W4313591085 hasConcept C177774035 @default.
- W4313591085 hasConcept C2776401178 @default.
- W4313591085 hasConcept C2780226545 @default.
- W4313591085 hasConcept C31972630 @default.
- W4313591085 hasConcept C41008148 @default.
- W4313591085 hasConcept C41895202 @default.
- W4313591085 hasConcept C52622490 @default.
- W4313591085 hasConcept C78201319 @default.
- W4313591085 hasConceptScore W4313591085C111919701 @default.
- W4313591085 hasConceptScore W4313591085C138885662 @default.
- W4313591085 hasConceptScore W4313591085C150899416 @default.
- W4313591085 hasConceptScore W4313591085C153180895 @default.
- W4313591085 hasConceptScore W4313591085C154945302 @default.
- W4313591085 hasConceptScore W4313591085C160633673 @default.
- W4313591085 hasConceptScore W4313591085C177774035 @default.
- W4313591085 hasConceptScore W4313591085C2776401178 @default.
- W4313591085 hasConceptScore W4313591085C2780226545 @default.
- W4313591085 hasConceptScore W4313591085C31972630 @default.
- W4313591085 hasConceptScore W4313591085C41008148 @default.