Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313307857> ?p ?o ?g. }
- W4313307857 endingPage "9037" @default.
- W4313307857 startingPage "9022" @default.
- W4313307857 abstract "Visible–thermal person reidentification (VT-ReID) is an emerging cross-modality matching problem, which aims to identify the same person across the daytime visible modality and nighttime thermal modality in the Internet of Things. Existing cutting-edge approaches consistently attempt to exploit image generation technique to generate cross-modality images or design various feature-level constraints to align feature distribution of heterogeneous data. However, color variations originating from the different imaging processes of spectrum cameras remain unsolved, which leads to suboptimal feature representations. In this article, we present a simple but very effective data augmentation method named Random Gray for the cross-modality matching task. Given a training sample, Random Gray randomly selects a rectangular region and translates it to grayscale. In this process, training images with fusing various levels of visible and grayscale information are generated, thereby reducing the risk of overfitting and making the model robust to color variations. Besides, we introduce a novel pooling method called softpooling to retain more information in the reduced activation maps. With softpooling layer, the network can learn more discriminative person features and further boost its retrieval performance. We conduct extensive experiments on publicly available cross-modality Re-ID data sets (SYSU-MM01 and RegDB) to demonstrate the effectiveness of our proposed method. Experimental results show that Random Gray and softpooling strategies yield significant accuracy improvement, and they can be utilized as training tricks for further VT-ReID research." @default.
- W4313307857 created "2023-01-06" @default.
- W4313307857 creator A5001943246 @default.
- W4313307857 creator A5026776669 @default.
- W4313307857 creator A5031916986 @default.
- W4313307857 creator A5035484030 @default.
- W4313307857 creator A5076624593 @default.
- W4313307857 date "2023-05-15" @default.
- W4313307857 modified "2023-09-25" @default.
- W4313307857 title "Visible–Thermal Person Reidentification in Visual Internet of Things With Random Gray Data Augmentation and a New Pooling Mechanism" @default.
- W4313307857 cites W166429404 @default.
- W4313307857 cites W2220271458 @default.
- W4313307857 cites W2502225121 @default.
- W4313307857 cites W2559156603 @default.
- W4313307857 cites W2596603442 @default.
- W4313307857 cites W2604463754 @default.
- W4313307857 cites W2766623491 @default.
- W4313307857 cites W2777534232 @default.
- W4313307857 cites W2788846526 @default.
- W4313307857 cites W2798429327 @default.
- W4313307857 cites W2798550112 @default.
- W4313307857 cites W2802152130 @default.
- W4313307857 cites W2808134123 @default.
- W4313307857 cites W2808260522 @default.
- W4313307857 cites W2888159112 @default.
- W4313307857 cites W2898047322 @default.
- W4313307857 cites W2954773727 @default.
- W4313307857 cites W2963597983 @default.
- W4313307857 cites W2970390221 @default.
- W4313307857 cites W2980073905 @default.
- W4313307857 cites W2980827638 @default.
- W4313307857 cites W2985033611 @default.
- W4313307857 cites W2996199246 @default.
- W4313307857 cites W2997052190 @default.
- W4313307857 cites W2997877744 @default.
- W4313307857 cites W2998508940 @default.
- W4313307857 cites W2998792609 @default.
- W4313307857 cites W3009246260 @default.
- W4313307857 cites W3026581226 @default.
- W4313307857 cites W3033235266 @default.
- W4313307857 cites W3034084848 @default.
- W4313307857 cites W3036401722 @default.
- W4313307857 cites W3047035309 @default.
- W4313307857 cites W3100555577 @default.
- W4313307857 cites W3102911434 @default.
- W4313307857 cites W3106715802 @default.
- W4313307857 cites W3111440907 @default.
- W4313307857 cites W3112063620 @default.
- W4313307857 cites W3134669182 @default.
- W4313307857 cites W3137802886 @default.
- W4313307857 cites W3139299097 @default.
- W4313307857 cites W3142867433 @default.
- W4313307857 cites W3176633985 @default.
- W4313307857 cites W3202788649 @default.
- W4313307857 cites W4224324173 @default.
- W4313307857 cites W46454230 @default.
- W4313307857 doi "https://doi.org/10.1109/jiot.2022.3233183" @default.
- W4313307857 hasPublicationYear "2023" @default.
- W4313307857 type Work @default.
- W4313307857 citedByCount "0" @default.
- W4313307857 crossrefType "journal-article" @default.
- W4313307857 hasAuthorship W4313307857A5001943246 @default.
- W4313307857 hasAuthorship W4313307857A5026776669 @default.
- W4313307857 hasAuthorship W4313307857A5031916986 @default.
- W4313307857 hasAuthorship W4313307857A5035484030 @default.
- W4313307857 hasAuthorship W4313307857A5076624593 @default.
- W4313307857 hasConcept C115961682 @default.
- W4313307857 hasConcept C119857082 @default.
- W4313307857 hasConcept C153180895 @default.
- W4313307857 hasConcept C154945302 @default.
- W4313307857 hasConcept C169258074 @default.
- W4313307857 hasConcept C22019652 @default.
- W4313307857 hasConcept C2780226545 @default.
- W4313307857 hasConcept C31972630 @default.
- W4313307857 hasConcept C41008148 @default.
- W4313307857 hasConcept C50644808 @default.
- W4313307857 hasConcept C70437156 @default.
- W4313307857 hasConcept C78201319 @default.
- W4313307857 hasConcept C97931131 @default.
- W4313307857 hasConceptScore W4313307857C115961682 @default.
- W4313307857 hasConceptScore W4313307857C119857082 @default.
- W4313307857 hasConceptScore W4313307857C153180895 @default.
- W4313307857 hasConceptScore W4313307857C154945302 @default.
- W4313307857 hasConceptScore W4313307857C169258074 @default.
- W4313307857 hasConceptScore W4313307857C22019652 @default.
- W4313307857 hasConceptScore W4313307857C2780226545 @default.
- W4313307857 hasConceptScore W4313307857C31972630 @default.
- W4313307857 hasConceptScore W4313307857C41008148 @default.
- W4313307857 hasConceptScore W4313307857C50644808 @default.
- W4313307857 hasConceptScore W4313307857C70437156 @default.
- W4313307857 hasConceptScore W4313307857C78201319 @default.
- W4313307857 hasConceptScore W4313307857C97931131 @default.
- W4313307857 hasFunder F4320321001 @default.
- W4313307857 hasIssue "10" @default.
- W4313307857 hasLocation W43133078571 @default.
- W4313307857 hasOpenAccess W4313307857 @default.
- W4313307857 hasPrimaryLocation W43133078571 @default.
- W4313307857 hasRelatedWork W2024160000 @default.