Matches in SemOpenAlex for { <https://semopenalex.org/work/W4361762922> ?p ?o ?g. }
- W4361762922 endingPage "143" @default.
- W4361762922 startingPage "131" @default.
- W4361762922 abstract "The past decade has seen the rapid development of deep learning methods in person re-identification. However, most existing models ignore the sharing and exclusive information between different images, leading to insufficient discriminative information from inter-class images. Therefore, this paper proposes a multi-branch graph convolution model to learn the sharing and exclusive information both at the global and local levels. First, using multi-granularity features as graph nodes and combining cosine similarity to construct local and global graphs to mine the relationships between pedestrian images; then, embedding the inter-local and global relationships into the feature representation of pedestrian images by graph convolution operations; finally, combining identity loss and component segmentation loss as the final loss function for the model training. Experimental results on the Market-1501, CUHK03, and DukeMTMC-reID datasets show that the proposed method can effectively improve the performance of person re-identification." @default.
- W4361762922 created "2023-04-04" @default.
- W4361762922 creator A5008875459 @default.
- W4361762922 creator A5014599311 @default.
- W4361762922 creator A5036130100 @default.
- W4361762922 creator A5074912702 @default.
- W4361762922 date "2023-01-01" @default.
- W4361762922 modified "2023-09-25" @default.
- W4361762922 title "Graph-Based Multi-granularity Person Re-identification" @default.
- W4361762922 cites W1982925187 @default.
- W4361762922 cites W2194775991 @default.
- W4361762922 cites W2204750386 @default.
- W4361762922 cites W2585635281 @default.
- W4361762922 cites W2604463754 @default.
- W4361762922 cites W2606202972 @default.
- W4361762922 cites W2798874329 @default.
- W4361762922 cites W2883638665 @default.
- W4361762922 cites W2946574625 @default.
- W4361762922 cites W2955058313 @default.
- W4361762922 cites W2963805953 @default.
- W4361762922 cites W2963876278 @default.
- W4361762922 cites W2979931389 @default.
- W4361762922 cites W2988964414 @default.
- W4361762922 cites W2992183111 @default.
- W4361762922 cites W3035252826 @default.
- W4361762922 cites W3037140251 @default.
- W4361762922 cites W3087658752 @default.
- W4361762922 cites W3100506510 @default.
- W4361762922 cites W3162562365 @default.
- W4361762922 cites W3174189418 @default.
- W4361762922 cites W3180036565 @default.
- W4361762922 doi "https://doi.org/10.1007/978-981-99-0923-0_14" @default.
- W4361762922 hasPublicationYear "2023" @default.
- W4361762922 type Work @default.
- W4361762922 citedByCount "0" @default.
- W4361762922 crossrefType "book-chapter" @default.
- W4361762922 hasAuthorship W4361762922A5008875459 @default.
- W4361762922 hasAuthorship W4361762922A5014599311 @default.
- W4361762922 hasAuthorship W4361762922A5036130100 @default.
- W4361762922 hasAuthorship W4361762922A5074912702 @default.
- W4361762922 hasConcept C111919701 @default.
- W4361762922 hasConcept C116834253 @default.
- W4361762922 hasConcept C124101348 @default.
- W4361762922 hasConcept C127413603 @default.
- W4361762922 hasConcept C132525143 @default.
- W4361762922 hasConcept C138885662 @default.
- W4361762922 hasConcept C153180895 @default.
- W4361762922 hasConcept C154945302 @default.
- W4361762922 hasConcept C177774035 @default.
- W4361762922 hasConcept C22212356 @default.
- W4361762922 hasConcept C2776401178 @default.
- W4361762922 hasConcept C2777113093 @default.
- W4361762922 hasConcept C2988416141 @default.
- W4361762922 hasConcept C41008148 @default.
- W4361762922 hasConcept C41608201 @default.
- W4361762922 hasConcept C41895202 @default.
- W4361762922 hasConcept C45347329 @default.
- W4361762922 hasConcept C50644808 @default.
- W4361762922 hasConcept C59404180 @default.
- W4361762922 hasConcept C59822182 @default.
- W4361762922 hasConcept C80444323 @default.
- W4361762922 hasConcept C86803240 @default.
- W4361762922 hasConcept C89600930 @default.
- W4361762922 hasConcept C97931131 @default.
- W4361762922 hasConceptScore W4361762922C111919701 @default.
- W4361762922 hasConceptScore W4361762922C116834253 @default.
- W4361762922 hasConceptScore W4361762922C124101348 @default.
- W4361762922 hasConceptScore W4361762922C127413603 @default.
- W4361762922 hasConceptScore W4361762922C132525143 @default.
- W4361762922 hasConceptScore W4361762922C138885662 @default.
- W4361762922 hasConceptScore W4361762922C153180895 @default.
- W4361762922 hasConceptScore W4361762922C154945302 @default.
- W4361762922 hasConceptScore W4361762922C177774035 @default.
- W4361762922 hasConceptScore W4361762922C22212356 @default.
- W4361762922 hasConceptScore W4361762922C2776401178 @default.
- W4361762922 hasConceptScore W4361762922C2777113093 @default.
- W4361762922 hasConceptScore W4361762922C2988416141 @default.
- W4361762922 hasConceptScore W4361762922C41008148 @default.
- W4361762922 hasConceptScore W4361762922C41608201 @default.
- W4361762922 hasConceptScore W4361762922C41895202 @default.
- W4361762922 hasConceptScore W4361762922C45347329 @default.
- W4361762922 hasConceptScore W4361762922C50644808 @default.
- W4361762922 hasConceptScore W4361762922C59404180 @default.
- W4361762922 hasConceptScore W4361762922C59822182 @default.
- W4361762922 hasConceptScore W4361762922C80444323 @default.
- W4361762922 hasConceptScore W4361762922C86803240 @default.
- W4361762922 hasConceptScore W4361762922C89600930 @default.
- W4361762922 hasConceptScore W4361762922C97931131 @default.
- W4361762922 hasLocation W43617629221 @default.
- W4361762922 hasOpenAccess W4361762922 @default.
- W4361762922 hasPrimaryLocation W43617629221 @default.
- W4361762922 hasRelatedWork W1971623867 @default.
- W4361762922 hasRelatedWork W2050806332 @default.
- W4361762922 hasRelatedWork W2061273563 @default.
- W4361762922 hasRelatedWork W2510758617 @default.
- W4361762922 hasRelatedWork W2546942002 @default.
- W4361762922 hasRelatedWork W2554403468 @default.
- W4361762922 hasRelatedWork W2573132524 @default.