Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313019229> ?p ?o ?g. }
- W4313019229 endingPage "628" @default.
- W4313019229 startingPage "612" @default.
- W4313019229 abstract "Weakly Supervised Object Localization (WSOL), which aims to localize objects by only using image-level labels, has attracted much attention because of its low annotation cost in real applications. Recent studies leverage the advantage of self-attention in visual Transformer for long-range dependency to re-active semantic regions, aiming to avoid partial activation in traditional class activation mapping (CAM). However, the long-range modeling in Transformer neglects the inherent spatial coherence of the object, and it usually diffuses the semantic-aware regions far from the object boundary, making localization results significantly larger or far smaller. To address such an issue, we introduce a simple yet effective Spatial Calibration Module (SCM) for accurate WSOL, incorporating semantic similarities of patch tokens and their spatial relationships into a unified diffusion model. Specifically, we introduce a learnable parameter to dynamically adjust the semantic correlations and spatial context intensities for effective information propagation. In practice, SCM is designed as an external module of Transformer, and can be removed during inference to reduce the computation cost. The object-sensitive localization ability is implicitly embedded into the Transformer encoder through optimization in the training phase. It enables the generated attention maps to capture the sharper object boundaries and filter the object-irrelevant background area. Extensive experimental results demonstrate the effectiveness of the proposed method, which significantly outperforms its counterpart TS-CAM on both CUB-200 and ImageNet-1K benchmarks. The code is available at https://github.com/164140757/SCM." @default.
- W4313019229 created "2023-01-05" @default.
- W4313019229 creator A5003608795 @default.
- W4313019229 creator A5030367905 @default.
- W4313019229 creator A5042137723 @default.
- W4313019229 creator A5083845568 @default.
- W4313019229 date "2022-01-01" @default.
- W4313019229 modified "2023-10-16" @default.
- W4313019229 title "Weakly Supervised Object Localization via Transformer with Implicit Spatial Calibration" @default.
- W4313019229 cites W1510646708 @default.
- W4313019229 cites W1677182931 @default.
- W4313019229 cites W1979036815 @default.
- W4313019229 cites W2052575990 @default.
- W4313019229 cites W2081010397 @default.
- W4313019229 cites W2095016473 @default.
- W4313019229 cites W2117539524 @default.
- W4313019229 cites W2129997546 @default.
- W4313019229 cites W2295107390 @default.
- W4313019229 cites W2883554151 @default.
- W4313019229 cites W2950557962 @default.
- W4313019229 cites W2963045696 @default.
- W4313019229 cites W2963108253 @default.
- W4313019229 cites W2963739249 @default.
- W4313019229 cites W2964274719 @default.
- W4313019229 cites W2990371274 @default.
- W4313019229 cites W2992308087 @default.
- W4313019229 cites W3034315787 @default.
- W4313019229 cites W3035318183 @default.
- W4313019229 cites W3035712793 @default.
- W4313019229 cites W3097777922 @default.
- W4313019229 cites W3107169861 @default.
- W4313019229 cites W3110272085 @default.
- W4313019229 cites W3138516171 @default.
- W4313019229 cites W3142837074 @default.
- W4313019229 cites W3188615767 @default.
- W4313019229 cites W4200631217 @default.
- W4313019229 cites W4210335518 @default.
- W4313019229 cites W4230938240 @default.
- W4313019229 cites W4312441698 @default.
- W4313019229 cites W4312570500 @default.
- W4313019229 doi "https://doi.org/10.1007/978-3-031-20077-9_36" @default.
- W4313019229 hasPublicationYear "2022" @default.
- W4313019229 type Work @default.
- W4313019229 citedByCount "4" @default.
- W4313019229 countsByYear W43130192292022 @default.
- W4313019229 countsByYear W43130192292023 @default.
- W4313019229 crossrefType "book-chapter" @default.
- W4313019229 hasAuthorship W4313019229A5003608795 @default.
- W4313019229 hasAuthorship W4313019229A5030367905 @default.
- W4313019229 hasAuthorship W4313019229A5042137723 @default.
- W4313019229 hasAuthorship W4313019229A5083845568 @default.
- W4313019229 hasBestOaLocation W43130192292 @default.
- W4313019229 hasConcept C111919701 @default.
- W4313019229 hasConcept C11413529 @default.
- W4313019229 hasConcept C118505674 @default.
- W4313019229 hasConcept C121332964 @default.
- W4313019229 hasConcept C153083717 @default.
- W4313019229 hasConcept C153180895 @default.
- W4313019229 hasConcept C154945302 @default.
- W4313019229 hasConcept C165801399 @default.
- W4313019229 hasConcept C2775955345 @default.
- W4313019229 hasConcept C2776214188 @default.
- W4313019229 hasConcept C31972630 @default.
- W4313019229 hasConcept C41008148 @default.
- W4313019229 hasConcept C45374587 @default.
- W4313019229 hasConcept C62520636 @default.
- W4313019229 hasConcept C66322947 @default.
- W4313019229 hasConceptScore W4313019229C111919701 @default.
- W4313019229 hasConceptScore W4313019229C11413529 @default.
- W4313019229 hasConceptScore W4313019229C118505674 @default.
- W4313019229 hasConceptScore W4313019229C121332964 @default.
- W4313019229 hasConceptScore W4313019229C153083717 @default.
- W4313019229 hasConceptScore W4313019229C153180895 @default.
- W4313019229 hasConceptScore W4313019229C154945302 @default.
- W4313019229 hasConceptScore W4313019229C165801399 @default.
- W4313019229 hasConceptScore W4313019229C2775955345 @default.
- W4313019229 hasConceptScore W4313019229C2776214188 @default.
- W4313019229 hasConceptScore W4313019229C31972630 @default.
- W4313019229 hasConceptScore W4313019229C41008148 @default.
- W4313019229 hasConceptScore W4313019229C45374587 @default.
- W4313019229 hasConceptScore W4313019229C62520636 @default.
- W4313019229 hasConceptScore W4313019229C66322947 @default.
- W4313019229 hasLocation W43130192291 @default.
- W4313019229 hasLocation W43130192292 @default.
- W4313019229 hasLocation W43130192293 @default.
- W4313019229 hasOpenAccess W4313019229 @default.
- W4313019229 hasPrimaryLocation W43130192291 @default.
- W4313019229 hasRelatedWork W1891287906 @default.
- W4313019229 hasRelatedWork W1969923398 @default.
- W4313019229 hasRelatedWork W2036807459 @default.
- W4313019229 hasRelatedWork W2166024367 @default.
- W4313019229 hasRelatedWork W2275988210 @default.
- W4313019229 hasRelatedWork W2772917594 @default.
- W4313019229 hasRelatedWork W2775347418 @default.
- W4313019229 hasRelatedWork W4320830014 @default.
- W4313019229 hasRelatedWork W4385825655 @default.
- W4313019229 hasRelatedWork W4386566506 @default.
- W4313019229 isParatext "false" @default.