Matches in SemOpenAlex for { <https://semopenalex.org/work/W3135717814> ?p ?o ?g. }
- W3135717814 abstract "Data augmentation is vital for deep learning neural networks. By providing massive training samples, it helps to improve the generalization ability of the model. Weakly supervised semantic segmentation (WSSS) is a challenging problem that has been deeply studied in recent years, conventional data augmentation approaches for WSSS usually employ geometrical transformations, random cropping and color jittering. However, merely increasing the same contextual semantic data does not bring much gain to the networks to distinguish the objects, e.g., the correct image-level classification of aeroplane may be not only due to the recognition of the object itself, but also its co-occurrence context like sky, which will cause the model to focus less on the object features. To this end, we present a Context Decoupling Augmentation (CDA) method, to change the inherent context in which the objects appear and thus drive the network to remove the dependence between object instances and contextual information. To validate the effectiveness of the proposed method, extensive experiments on PASCAL VOC 2012 dataset with several alternative network architectures demonstrate that CDA can boost various popular WSSS methods to the new state-of-the-art by a large margin." @default.
- W3135717814 created "2021-03-15" @default.
- W3135717814 creator A5003812660 @default.
- W3135717814 creator A5023130798 @default.
- W3135717814 creator A5029912845 @default.
- W3135717814 creator A5029998379 @default.
- W3135717814 date "2021-03-02" @default.
- W3135717814 modified "2023-09-27" @default.
- W3135717814 title "Context Decoupling Augmentation for Weakly Supervised Semantic Segmentation" @default.
- W3135717814 cites W1495267108 @default.
- W3135717814 cites W1783315696 @default.
- W3135717814 cites W1861492603 @default.
- W3135717814 cites W1903029394 @default.
- W3135717814 cites W2037227137 @default.
- W3135717814 cites W2108598243 @default.
- W3135717814 cites W2144794286 @default.
- W3135717814 cites W2161236525 @default.
- W3135717814 cites W2194775991 @default.
- W3135717814 cites W2306289963 @default.
- W3135717814 cites W2337429362 @default.
- W3135717814 cites W2412782625 @default.
- W3135717814 cites W2516803306 @default.
- W3135717814 cites W2552414813 @default.
- W3135717814 cites W2585747585 @default.
- W3135717814 cites W2600144439 @default.
- W3135717814 cites W2746314669 @default.
- W3135717814 cites W2765407302 @default.
- W3135717814 cites W2798376494 @default.
- W3135717814 cites W2799124825 @default.
- W3135717814 cites W2883820570 @default.
- W3135717814 cites W2946952476 @default.
- W3135717814 cites W2952610664 @default.
- W3135717814 cites W2962758679 @default.
- W3135717814 cites W2962759496 @default.
- W3135717814 cites W2962793481 @default.
- W3135717814 cites W2962867364 @default.
- W3135717814 cites W2963237621 @default.
- W3135717814 cites W2963271314 @default.
- W3135717814 cites W2963346885 @default.
- W3135717814 cites W2963458902 @default.
- W3135717814 cites W2963516811 @default.
- W3135717814 cites W2963563573 @default.
- W3135717814 cites W2963606198 @default.
- W3135717814 cites W2963840672 @default.
- W3135717814 cites W2963942586 @default.
- W3135717814 cites W2963977581 @default.
- W3135717814 cites W2964121744 @default.
- W3135717814 cites W2964288706 @default.
- W3135717814 cites W2964309882 @default.
- W3135717814 cites W2980189057 @default.
- W3135717814 cites W2992308087 @default.
- W3135717814 cites W3034333089 @default.
- W3135717814 cites W3034930876 @default.
- W3135717814 cites W3098191145 @default.
- W3135717814 cites W3100040694 @default.
- W3135717814 cites W3104205547 @default.
- W3135717814 cites W611457968 @default.
- W3135717814 cites W639708223 @default.
- W3135717814 doi "https://doi.org/10.48550/arxiv.2103.01795" @default.
- W3135717814 hasPublicationYear "2021" @default.
- W3135717814 type Work @default.
- W3135717814 sameAs 3135717814 @default.
- W3135717814 citedByCount "0" @default.
- W3135717814 crossrefType "posted-content" @default.
- W3135717814 hasAuthorship W3135717814A5003812660 @default.
- W3135717814 hasAuthorship W3135717814A5023130798 @default.
- W3135717814 hasAuthorship W3135717814A5029912845 @default.
- W3135717814 hasAuthorship W3135717814A5029998379 @default.
- W3135717814 hasBestOaLocation W31357178141 @default.
- W3135717814 hasConcept C119857082 @default.
- W3135717814 hasConcept C127413603 @default.
- W3135717814 hasConcept C133731056 @default.
- W3135717814 hasConcept C151730666 @default.
- W3135717814 hasConcept C153180895 @default.
- W3135717814 hasConcept C154945302 @default.
- W3135717814 hasConcept C199360897 @default.
- W3135717814 hasConcept C205606062 @default.
- W3135717814 hasConcept C2779343474 @default.
- W3135717814 hasConcept C41008148 @default.
- W3135717814 hasConcept C50644808 @default.
- W3135717814 hasConcept C75608658 @default.
- W3135717814 hasConcept C774472 @default.
- W3135717814 hasConcept C86803240 @default.
- W3135717814 hasConcept C89600930 @default.
- W3135717814 hasConceptScore W3135717814C119857082 @default.
- W3135717814 hasConceptScore W3135717814C127413603 @default.
- W3135717814 hasConceptScore W3135717814C133731056 @default.
- W3135717814 hasConceptScore W3135717814C151730666 @default.
- W3135717814 hasConceptScore W3135717814C153180895 @default.
- W3135717814 hasConceptScore W3135717814C154945302 @default.
- W3135717814 hasConceptScore W3135717814C199360897 @default.
- W3135717814 hasConceptScore W3135717814C205606062 @default.
- W3135717814 hasConceptScore W3135717814C2779343474 @default.
- W3135717814 hasConceptScore W3135717814C41008148 @default.
- W3135717814 hasConceptScore W3135717814C50644808 @default.
- W3135717814 hasConceptScore W3135717814C75608658 @default.
- W3135717814 hasConceptScore W3135717814C774472 @default.
- W3135717814 hasConceptScore W3135717814C86803240 @default.
- W3135717814 hasConceptScore W3135717814C89600930 @default.
- W3135717814 hasLocation W31357178141 @default.