Matches in SemOpenAlex for { <https://semopenalex.org/work/W3118536434> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W3118536434 endingPage "15199" @default.
- W3118536434 startingPage "15191" @default.
- W3118536434 abstract "Data augmentation is an effective way to increase the diversity of existing training datasets that result in improved generalization ability of convolutional neural networks (CNNs). The augmentation effect is usually global for the existing methods i.e., a single augmentation effect is applied to the whole image, thus limiting the diversity of local characteristics in augmented images. Moreover, the global augmentation effect does not support the most fundamental behavior of CNNs i.e., they focus more on local features (local texture, tiny noise etc.) than global shapes. We refer to this behavior as local bias property. In this paper, we propose a new data augmentation method, called Local Augment (LA), which highly alters the local bias property so that it can generate significantly diverse augmented images and offers the network with a better augmentation effect. First, we select few local patches in an image, then apply different types of augmentation strategies to each local patch. This augmentation process collapses the global structure of the object but creates locally diversified samples, which helps the network to learn the local bias property in a more generalized way. As a result, it increases the generalizability and the prediction accuracy of the network. To verify the effectiveness of the proposed method, we perform comprehensive experiments on image classification with benchmark datasets, where the proposed method outperforms the sate-of-the-art data augmentation techniques on ImageNet and STL10 and shows competitive performance on CIFAR100." @default.
- W3118536434 created "2021-01-18" @default.
- W3118536434 creator A5005200603 @default.
- W3118536434 creator A5022958958 @default.
- W3118536434 creator A5040139490 @default.
- W3118536434 date "2021-01-01" @default.
- W3118536434 modified "2023-10-16" @default.
- W3118536434 title "Local Augment: Utilizing Local Bias Property of Convolutional Neural Networks for Data Augmentation" @default.
- W3118536434 cites W1903029394 @default.
- W3118536434 cites W2117539524 @default.
- W3118536434 cites W2194775991 @default.
- W3118536434 cites W2295107390 @default.
- W3118536434 cites W2412782625 @default.
- W3118536434 cites W2549139847 @default.
- W3118536434 cites W2603777577 @default.
- W3118536434 cites W2903867357 @default.
- W3118536434 cites W2963622428 @default.
- W3118536434 cites W2992308087 @default.
- W3118536434 cites W2998508940 @default.
- W3118536434 cites W3106250896 @default.
- W3118536434 doi "https://doi.org/10.1109/access.2021.3050758" @default.
- W3118536434 hasPublicationYear "2021" @default.
- W3118536434 type Work @default.
- W3118536434 sameAs 3118536434 @default.
- W3118536434 citedByCount "8" @default.
- W3118536434 countsByYear W31185364342021 @default.
- W3118536434 countsByYear W31185364342022 @default.
- W3118536434 countsByYear W31185364342023 @default.
- W3118536434 crossrefType "journal-article" @default.
- W3118536434 hasAuthorship W3118536434A5005200603 @default.
- W3118536434 hasAuthorship W3118536434A5022958958 @default.
- W3118536434 hasAuthorship W3118536434A5040139490 @default.
- W3118536434 hasBestOaLocation W31185364341 @default.
- W3118536434 hasConcept C105795698 @default.
- W3118536434 hasConcept C111472728 @default.
- W3118536434 hasConcept C115961682 @default.
- W3118536434 hasConcept C119857082 @default.
- W3118536434 hasConcept C13280743 @default.
- W3118536434 hasConcept C134306372 @default.
- W3118536434 hasConcept C138885662 @default.
- W3118536434 hasConcept C153180895 @default.
- W3118536434 hasConcept C154945302 @default.
- W3118536434 hasConcept C177148314 @default.
- W3118536434 hasConcept C185798385 @default.
- W3118536434 hasConcept C189950617 @default.
- W3118536434 hasConcept C205649164 @default.
- W3118536434 hasConcept C27158222 @default.
- W3118536434 hasConcept C33923547 @default.
- W3118536434 hasConcept C41008148 @default.
- W3118536434 hasConcept C50644808 @default.
- W3118536434 hasConcept C81363708 @default.
- W3118536434 hasConceptScore W3118536434C105795698 @default.
- W3118536434 hasConceptScore W3118536434C111472728 @default.
- W3118536434 hasConceptScore W3118536434C115961682 @default.
- W3118536434 hasConceptScore W3118536434C119857082 @default.
- W3118536434 hasConceptScore W3118536434C13280743 @default.
- W3118536434 hasConceptScore W3118536434C134306372 @default.
- W3118536434 hasConceptScore W3118536434C138885662 @default.
- W3118536434 hasConceptScore W3118536434C153180895 @default.
- W3118536434 hasConceptScore W3118536434C154945302 @default.
- W3118536434 hasConceptScore W3118536434C177148314 @default.
- W3118536434 hasConceptScore W3118536434C185798385 @default.
- W3118536434 hasConceptScore W3118536434C189950617 @default.
- W3118536434 hasConceptScore W3118536434C205649164 @default.
- W3118536434 hasConceptScore W3118536434C27158222 @default.
- W3118536434 hasConceptScore W3118536434C33923547 @default.
- W3118536434 hasConceptScore W3118536434C41008148 @default.
- W3118536434 hasConceptScore W3118536434C50644808 @default.
- W3118536434 hasConceptScore W3118536434C81363708 @default.
- W3118536434 hasFunder F4320335489 @default.
- W3118536434 hasLocation W31185364341 @default.
- W3118536434 hasLocation W31185364342 @default.
- W3118536434 hasOpenAccess W3118536434 @default.
- W3118536434 hasPrimaryLocation W31185364341 @default.
- W3118536434 hasRelatedWork W2767651786 @default.
- W3118536434 hasRelatedWork W2912288872 @default.
- W3118536434 hasRelatedWork W3021430260 @default.
- W3118536434 hasRelatedWork W3027997911 @default.
- W3118536434 hasRelatedWork W4221150964 @default.
- W3118536434 hasRelatedWork W4287776258 @default.
- W3118536434 hasRelatedWork W4312365371 @default.
- W3118536434 hasRelatedWork W4379251595 @default.
- W3118536434 hasRelatedWork W4385189843 @default.
- W3118536434 hasRelatedWork W564581980 @default.
- W3118536434 hasVolume "9" @default.
- W3118536434 isParatext "false" @default.
- W3118536434 isRetracted "false" @default.
- W3118536434 magId "3118536434" @default.
- W3118536434 workType "article" @default.