Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387185524> ?p ?o ?g. }
Showing items 1 to 59 of
59
with 100 items per page.
- W4387185524 abstract "Image-level weakly supervised semantic segmentation (WSSS) has attracted much attention due to the easily acquired class labels. Most existing methods resort to utilizing Class Activation Maps (CAMs) obtained from the classification network to play as the initial pseudo labels. However, the classifiers only focus on the most discriminative regions of the target objects, which is referred to as the information bottleneck from the perspective of the information theory. To alleviate this information bottleneck limitation, we propose an Information Perturbation Module (IPM) to explicitly obtain the information difference maps, which provide the accurate direction and magnitude of the information compression in the classification network. After that, an information bottleneck breakthrough mechanism with three branches is proposed to overcome the information bottleneck in the classification network for segmentation. Additionally, a diversity regularization on the generated two information difference maps is proposed to improve the diversity of the output CAMs. Extensive experiments on PASCAL VOC2012 val and test sets demonstrate that the proposed method can effectively improve the weakly supervised semantic segmentation performance of the advanced approaches." @default.
- W4387185524 created "2023-09-30" @default.
- W4387185524 creator A5026151620 @default.
- W4387185524 creator A5053437240 @default.
- W4387185524 creator A5061215231 @default.
- W4387185524 date "2023-09-28" @default.
- W4387185524 modified "2023-10-16" @default.
- W4387185524 title "Exploring Information Bottleneck for Weakly Supervised Semantic Segmentation" @default.
- W4387185524 doi "https://doi.org/10.3233/faia230480" @default.
- W4387185524 hasPublicationYear "2023" @default.
- W4387185524 type Work @default.
- W4387185524 citedByCount "0" @default.
- W4387185524 crossrefType "book-chapter" @default.
- W4387185524 hasAuthorship W4387185524A5026151620 @default.
- W4387185524 hasAuthorship W4387185524A5053437240 @default.
- W4387185524 hasAuthorship W4387185524A5061215231 @default.
- W4387185524 hasBestOaLocation W43871855241 @default.
- W4387185524 hasConcept C119857082 @default.
- W4387185524 hasConcept C124101348 @default.
- W4387185524 hasConcept C149635348 @default.
- W4387185524 hasConcept C153180895 @default.
- W4387185524 hasConcept C154945302 @default.
- W4387185524 hasConcept C199360897 @default.
- W4387185524 hasConcept C2780513914 @default.
- W4387185524 hasConcept C41008148 @default.
- W4387185524 hasConcept C60008888 @default.
- W4387185524 hasConcept C73555534 @default.
- W4387185524 hasConcept C75608658 @default.
- W4387185524 hasConcept C89600930 @default.
- W4387185524 hasConcept C97931131 @default.
- W4387185524 hasConceptScore W4387185524C119857082 @default.
- W4387185524 hasConceptScore W4387185524C124101348 @default.
- W4387185524 hasConceptScore W4387185524C149635348 @default.
- W4387185524 hasConceptScore W4387185524C153180895 @default.
- W4387185524 hasConceptScore W4387185524C154945302 @default.
- W4387185524 hasConceptScore W4387185524C199360897 @default.
- W4387185524 hasConceptScore W4387185524C2780513914 @default.
- W4387185524 hasConceptScore W4387185524C41008148 @default.
- W4387185524 hasConceptScore W4387185524C60008888 @default.
- W4387185524 hasConceptScore W4387185524C73555534 @default.
- W4387185524 hasConceptScore W4387185524C75608658 @default.
- W4387185524 hasConceptScore W4387185524C89600930 @default.
- W4387185524 hasConceptScore W4387185524C97931131 @default.
- W4387185524 hasLocation W43871855241 @default.
- W4387185524 hasOpenAccess W4387185524 @default.
- W4387185524 hasPrimaryLocation W43871855241 @default.
- W4387185524 hasRelatedWork W1972656095 @default.
- W4387185524 hasRelatedWork W2024160000 @default.
- W4387185524 hasRelatedWork W2061273563 @default.
- W4387185524 hasRelatedWork W2285052147 @default.
- W4387185524 hasRelatedWork W2729514902 @default.
- W4387185524 hasRelatedWork W2743258233 @default.
- W4387185524 hasRelatedWork W2773500201 @default.
- W4387185524 hasRelatedWork W2970216048 @default.
- W4387185524 hasRelatedWork W2998168123 @default.
- W4387185524 hasRelatedWork W4287995534 @default.
- W4387185524 isParatext "false" @default.
- W4387185524 isRetracted "false" @default.
- W4387185524 workType "book-chapter" @default.