Matches in SemOpenAlex for { <https://semopenalex.org/work/W3141336300> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W3141336300 abstract "With the rapid growth of volumes of unstructured data in the Internet, it is increasingly important to develop image-based search methods in the field of information retrieval. Hashing methods have been widely studied in favor of the low storage cost and fast retrieval speed. However, traditional hashing methods generate binary codes from handcrafted features, which limited its accuracy since the handcrafted features is hard to represent the semantic feature of the image. Recent studies have indicated that deep leaning models are superior in terms of feature extraction. Hence, in this paper, we employ the transfer learning strategy with attention mechanism on the extraction of semantic features and using difference hash algorithm to generate binary codes for image retrieval. In addition, the pre-trained CNN models also show good performance in semantic feature extraction. Instead of using the last layer of the pre-trained CNN model directly, we can provide a guidance on how to select semantic features in the task of image retrieval, by observing some changes between mean average precision and semantic features extracted from the pre-trained CNN model. Experimental results demonstrate that the mean average precision of proposed model achieves 0.944, 0.568 and 0.310 on the MNIST, Cifar-10 and Cifar-100 dataset respectively, which indicates that the transfer learning strategy with attention mechanism is effective in image retrieval." @default.
- W3141336300 created "2021-04-13" @default.
- W3141336300 creator A5022686531 @default.
- W3141336300 creator A5073989620 @default.
- W3141336300 creator A5084438133 @default.
- W3141336300 date "2021-01-01" @default.
- W3141336300 modified "2023-10-16" @default.
- W3141336300 title "Deep Semantic Hashing for Large-Scale Image Retrieval" @default.
- W3141336300 cites W1849277567 @default.
- W3141336300 cites W1915485278 @default.
- W3141336300 cites W2037651268 @default.
- W3141336300 cites W2082453965 @default.
- W3141336300 cites W2112796928 @default.
- W3141336300 cites W2123229215 @default.
- W3141336300 cites W2130660124 @default.
- W3141336300 cites W2464915613 @default.
- W3141336300 cites W2752782242 @default.
- W3141336300 cites W2955058313 @default.
- W3141336300 cites W2964095005 @default.
- W3141336300 cites W2972621596 @default.
- W3141336300 cites W2980275434 @default.
- W3141336300 doi "https://doi.org/10.1007/978-981-16-0479-9_13" @default.
- W3141336300 hasPublicationYear "2021" @default.
- W3141336300 type Work @default.
- W3141336300 sameAs 3141336300 @default.
- W3141336300 citedByCount "0" @default.
- W3141336300 crossrefType "book-chapter" @default.
- W3141336300 hasAuthorship W3141336300A5022686531 @default.
- W3141336300 hasAuthorship W3141336300A5073989620 @default.
- W3141336300 hasAuthorship W3141336300A5084438133 @default.
- W3141336300 hasConcept C108583219 @default.
- W3141336300 hasConcept C115961682 @default.
- W3141336300 hasConcept C138885662 @default.
- W3141336300 hasConcept C153180895 @default.
- W3141336300 hasConcept C154945302 @default.
- W3141336300 hasConcept C1667742 @default.
- W3141336300 hasConcept C190502265 @default.
- W3141336300 hasConcept C23123220 @default.
- W3141336300 hasConcept C2776401178 @default.
- W3141336300 hasConcept C2781122975 @default.
- W3141336300 hasConcept C33923547 @default.
- W3141336300 hasConcept C38652104 @default.
- W3141336300 hasConcept C41008148 @default.
- W3141336300 hasConcept C41895202 @default.
- W3141336300 hasConcept C48372109 @default.
- W3141336300 hasConcept C52622490 @default.
- W3141336300 hasConcept C63435697 @default.
- W3141336300 hasConcept C94375191 @default.
- W3141336300 hasConcept C99138194 @default.
- W3141336300 hasConceptScore W3141336300C108583219 @default.
- W3141336300 hasConceptScore W3141336300C115961682 @default.
- W3141336300 hasConceptScore W3141336300C138885662 @default.
- W3141336300 hasConceptScore W3141336300C153180895 @default.
- W3141336300 hasConceptScore W3141336300C154945302 @default.
- W3141336300 hasConceptScore W3141336300C1667742 @default.
- W3141336300 hasConceptScore W3141336300C190502265 @default.
- W3141336300 hasConceptScore W3141336300C23123220 @default.
- W3141336300 hasConceptScore W3141336300C2776401178 @default.
- W3141336300 hasConceptScore W3141336300C2781122975 @default.
- W3141336300 hasConceptScore W3141336300C33923547 @default.
- W3141336300 hasConceptScore W3141336300C38652104 @default.
- W3141336300 hasConceptScore W3141336300C41008148 @default.
- W3141336300 hasConceptScore W3141336300C41895202 @default.
- W3141336300 hasConceptScore W3141336300C48372109 @default.
- W3141336300 hasConceptScore W3141336300C52622490 @default.
- W3141336300 hasConceptScore W3141336300C63435697 @default.
- W3141336300 hasConceptScore W3141336300C94375191 @default.
- W3141336300 hasConceptScore W3141336300C99138194 @default.
- W3141336300 hasLocation W31413363001 @default.
- W3141336300 hasOpenAccess W3141336300 @default.
- W3141336300 hasPrimaryLocation W31413363001 @default.
- W3141336300 hasRelatedWork W10202958 @default.
- W3141336300 hasRelatedWork W11140586 @default.
- W3141336300 hasRelatedWork W12901231 @default.
- W3141336300 hasRelatedWork W1726589 @default.
- W3141336300 hasRelatedWork W3904675 @default.
- W3141336300 hasRelatedWork W6316837 @default.
- W3141336300 hasRelatedWork W7237889 @default.
- W3141336300 hasRelatedWork W9362070 @default.
- W3141336300 hasRelatedWork W9664128 @default.
- W3141336300 hasRelatedWork W967906 @default.
- W3141336300 isParatext "false" @default.
- W3141336300 isRetracted "false" @default.
- W3141336300 magId "3141336300" @default.
- W3141336300 workType "book-chapter" @default.