FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386913488> ?p ?o ?g. }

• W4386913488 endingPage "126764" @default.
• W4386913488 startingPage "126764" @default.
• W4386913488 abstract "Hashing has received broad attention in large-scale image retrieval due to its appealing efficiency in computation and storage. Particularly, with the drawn of deep learning, much efforts have been directed towards using deep neural networks to learn feature representations and hash codes simultaneously, and the developed deep hashing methods have shown superior performance over conventional hashing methods. In this paper, we propose Deep Attention Sampling Hashing (DASH), a novel deep hashing method that yields high-quality hash codes to enable efficient image retrieval. Specifically, we employ two sub-networks in DASH, i.e., a master branch and a part branch, to capture global structure features and discriminative feature representations, respectively. Furthermore, we develop an Attention Sampler Module (ASM), which consists of an Object Region Extraction (ORE) block and an Informative Patch Generation (IPG) block, to yield richer informative image patches. The ORE block provides a well-designed multi-scale attentional fusion mechanism to highlight and extract the significant regions of images, and the IPG block employs a direction-specific shift mechanism to generate desired image patches with discriminative details. Both blocks could be seamlessly integrated into various convolutional neural network (CNN) architectures. Subsequently, we conduct knowledge distillation optimization to transfer the details learned by the part branch into the master branch to guide hash code learning. In addition, we design a Weibull quantization loss to minimize the information loss caused by binary quantization. The experimental results on three benchmark datasets demonstrate the effectiveness of the proposed DASH with respect to different evaluation metrics." @default.
• W4386913488 created "2023-09-22" @default.
• W4386913488 creator A5015868975 @default.
• W4386913488 creator A5050683592 @default.
• W4386913488 creator A5068201443 @default.
• W4386913488 creator A5070818899 @default.
• W4386913488 date "2023-11-01" @default.
• W4386913488 modified "2023-10-06" @default.
• W4386913488 title "Deep attention sampling hashing for efficient image retrieval" @default.
• W4386913488 cites W1910300841 @default.
• W4386913488 cites W1939575207 @default.
• W4386913488 cites W1974647172 @default.
• W4386913488 cites W1992371516 @default.
• W4386913488 cites W2007972815 @default.
• W4386913488 cites W2012833704 @default.
• W4386913488 cites W2126210882 @default.
• W4386913488 cites W2162006472 @default.
• W4386913488 cites W2293824885 @default.
• W4386913488 cites W2411707397 @default.
• W4386913488 cites W2470673105 @default.
• W4386913488 cites W2508837377 @default.
• W4386913488 cites W2562322388 @default.
• W4386913488 cites W2604880013 @default.
• W4386913488 cites W2620998106 @default.
• W4386913488 cites W2738649458 @default.
• W4386913488 cites W2740912563 @default.
• W4386913488 cites W2752782242 @default.
• W4386913488 cites W2798834175 @default.
• W4386913488 cites W2798840456 @default.
• W4386913488 cites W2800282192 @default.
• W4386913488 cites W2801086478 @default.
• W4386913488 cites W2901281242 @default.
• W4386913488 cites W2902713587 @default.
• W4386913488 cites W2905097026 @default.
• W4386913488 cites W2905957315 @default.
• W4386913488 cites W2922521335 @default.
• W4386913488 cites W2948171447 @default.
• W4386913488 cites W2948671658 @default.
• W4386913488 cites W2951952490 @default.
• W4386913488 cites W2952787292 @default.
• W4386913488 cites W2954054736 @default.
• W4386913488 cites W2955058313 @default.
• W4386913488 cites W2963091558 @default.
• W4386913488 cites W2963140444 @default.
• W4386913488 cites W2963180826 @default.
• W4386913488 cites W2963407932 @default.
• W4386913488 cites W2964280870 @default.
• W4386913488 cites W2966380432 @default.
• W4386913488 cites W2992352554 @default.
• W4386913488 cites W3006683596 @default.
• W4386913488 cites W3007442642 @default.
• W4386913488 cites W3011666149 @default.
• W4386913488 cites W3021893287 @default.
• W4386913488 cites W3033453416 @default.
• W4386913488 cites W3034315422 @default.
• W4386913488 cites W3088032605 @default.
• W4386913488 cites W3094897602 @default.
• W4386913488 cites W3096806833 @default.
• W4386913488 cites W3100506510 @default.
• W4386913488 cites W3126558081 @default.
• W4386913488 cites W3127323395 @default.
• W4386913488 cites W3159460504 @default.
• W4386913488 cites W3203778378 @default.
• W4386913488 cites W4321636913 @default.
• W4386913488 doi "https://doi.org/10.1016/j.neucom.2023.126764" @default.
• W4386913488 hasPublicationYear "2023" @default.
• W4386913488 type Work @default.
• W4386913488 citedByCount "0" @default.
• W4386913488 crossrefType "journal-article" @default.
• W4386913488 hasAuthorship W4386913488A5015868975 @default.
• W4386913488 hasAuthorship W4386913488A5050683592 @default.
• W4386913488 hasAuthorship W4386913488A5068201443 @default.
• W4386913488 hasAuthorship W4386913488A5070818899 @default.
• W4386913488 hasConcept C108583219 @default.
• W4386913488 hasConcept C11413529 @default.
• W4386913488 hasConcept C115961682 @default.
• W4386913488 hasConcept C119857082 @default.
• W4386913488 hasConcept C133667856 @default.
• W4386913488 hasConcept C138111711 @default.
• W4386913488 hasConcept C153180895 @default.
• W4386913488 hasConcept C154945302 @default.
• W4386913488 hasConcept C1667742 @default.
• W4386913488 hasConcept C2524010 @default.
• W4386913488 hasConcept C2777210771 @default.
• W4386913488 hasConcept C28855332 @default.
• W4386913488 hasConcept C33923547 @default.
• W4386913488 hasConcept C38652104 @default.
• W4386913488 hasConcept C41008148 @default.
• W4386913488 hasConcept C48372109 @default.
• W4386913488 hasConcept C63435697 @default.
• W4386913488 hasConcept C67388219 @default.
• W4386913488 hasConcept C81363708 @default.
• W4386913488 hasConcept C94375191 @default.
• W4386913488 hasConcept C97931131 @default.
• W4386913488 hasConcept C99138194 @default.
• W4386913488 hasConceptScore W4386913488C108583219 @default.
• W4386913488 hasConceptScore W4386913488C11413529 @default.
• W4386913488 hasConceptScore W4386913488C115961682 @default.

• next