Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387130492> ?p ?o ?g. }
Showing items 1 to 74 of
74
with 100 items per page.
- W4387130492 abstract "Transformer network has been widely applied in the field of computer vision. Thanks to the application of self-attention mechanism, Transformer can extract and pay attention to features at each position in the image, capturing details more accurately. In this paper, in order to address the issue of obtaining different levels of classification results based on different semantic contexts, we propose a hierarchical image fine-grained classification network based on Swin Transformer. Specifically, multiple classification results of different classification granularity levels can be obtained for the same image. We also suggest a parallel multi-level feature mining module, establish a progressive training model, and use a patch generator to build training samples of different granularity and stages, so as to extract features of different classification granularity levels. We design a hierarchical feature fusion module to promote the classification results of the fine-grained branch on the coarse-grained branch. Furthermore, we present a weakly supervised feature selector to enhance the feature representation ability and improve the classification accuracy, which helps to select effective feature points from different stages of feature maps. Finally, we evaluate its performance on three benchmark datasets, and find that the proposed method achieves good performance in terms of weighted average accuracy." @default.
- W4387130492 created "2023-09-29" @default.
- W4387130492 creator A5010877224 @default.
- W4387130492 creator A5037831361 @default.
- W4387130492 creator A5064403648 @default.
- W4387130492 creator A5077171997 @default.
- W4387130492 creator A5081727539 @default.
- W4387130492 date "2023-07-27" @default.
- W4387130492 modified "2023-10-18" @default.
- W4387130492 title "Hierarchical Image Fine-Grained classification via Hierarchical Feature Mining and Filtering" @default.
- W4387130492 cites W2763070548 @default.
- W4387130492 cites W2765268259 @default.
- W4387130492 cites W3008809756 @default.
- W4387130492 cites W3105979354 @default.
- W4387130492 cites W3108870912 @default.
- W4387130492 cites W3150272974 @default.
- W4387130492 doi "https://doi.org/10.1145/3613330.3613334" @default.
- W4387130492 hasPublicationYear "2023" @default.
- W4387130492 type Work @default.
- W4387130492 citedByCount "0" @default.
- W4387130492 crossrefType "proceedings-article" @default.
- W4387130492 hasAuthorship W4387130492A5010877224 @default.
- W4387130492 hasAuthorship W4387130492A5037831361 @default.
- W4387130492 hasAuthorship W4387130492A5064403648 @default.
- W4387130492 hasAuthorship W4387130492A5077171997 @default.
- W4387130492 hasAuthorship W4387130492A5081727539 @default.
- W4387130492 hasConcept C111919701 @default.
- W4387130492 hasConcept C115961682 @default.
- W4387130492 hasConcept C119857082 @default.
- W4387130492 hasConcept C124101348 @default.
- W4387130492 hasConcept C13280743 @default.
- W4387130492 hasConcept C138885662 @default.
- W4387130492 hasConcept C153180895 @default.
- W4387130492 hasConcept C154945302 @default.
- W4387130492 hasConcept C177774035 @default.
- W4387130492 hasConcept C185798385 @default.
- W4387130492 hasConcept C205649164 @default.
- W4387130492 hasConcept C2776401178 @default.
- W4387130492 hasConcept C41008148 @default.
- W4387130492 hasConcept C41895202 @default.
- W4387130492 hasConcept C52622490 @default.
- W4387130492 hasConcept C75294576 @default.
- W4387130492 hasConceptScore W4387130492C111919701 @default.
- W4387130492 hasConceptScore W4387130492C115961682 @default.
- W4387130492 hasConceptScore W4387130492C119857082 @default.
- W4387130492 hasConceptScore W4387130492C124101348 @default.
- W4387130492 hasConceptScore W4387130492C13280743 @default.
- W4387130492 hasConceptScore W4387130492C138885662 @default.
- W4387130492 hasConceptScore W4387130492C153180895 @default.
- W4387130492 hasConceptScore W4387130492C154945302 @default.
- W4387130492 hasConceptScore W4387130492C177774035 @default.
- W4387130492 hasConceptScore W4387130492C185798385 @default.
- W4387130492 hasConceptScore W4387130492C205649164 @default.
- W4387130492 hasConceptScore W4387130492C2776401178 @default.
- W4387130492 hasConceptScore W4387130492C41008148 @default.
- W4387130492 hasConceptScore W4387130492C41895202 @default.
- W4387130492 hasConceptScore W4387130492C52622490 @default.
- W4387130492 hasConceptScore W4387130492C75294576 @default.
- W4387130492 hasLocation W43871304921 @default.
- W4387130492 hasOpenAccess W4387130492 @default.
- W4387130492 hasPrimaryLocation W43871304921 @default.
- W4387130492 hasRelatedWork W2144059113 @default.
- W4387130492 hasRelatedWork W2146076056 @default.
- W4387130492 hasRelatedWork W2146204105 @default.
- W4387130492 hasRelatedWork W2546942002 @default.
- W4387130492 hasRelatedWork W2738461075 @default.
- W4387130492 hasRelatedWork W2811390910 @default.
- W4387130492 hasRelatedWork W2940977206 @default.
- W4387130492 hasRelatedWork W3005023910 @default.
- W4387130492 hasRelatedWork W4288388940 @default.
- W4387130492 hasRelatedWork W4327773867 @default.
- W4387130492 isParatext "false" @default.
- W4387130492 isRetracted "false" @default.
- W4387130492 workType "article" @default.