Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386597013> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W4386597013 abstract "Deep convolutional neural networks (CNNs) have achieved remarkable success in computer vision tasks, but their training is susceptible to overfitting when the training sample size is insufficient. In this paper, we introduce Corner CutMix, a novel data augmentation technique for CNN training. During training, Corner CutMix randomly selects a region from one of four corner areas in an image and replaces it with a randomly chosen region from a distractor image. Additionally, we design an auxiliary self-supervised loss function to learn the position of the selected corner region, thereby improving the transferability and generalizability of the learned representation. Corner CutMix is easy to implement, adding little computational overhead, and can be combined with other augmentation methods such as random cropping, color distortion, and flipping. Our extensive classification task experiments in self-supervised learning on public datasets (e.g., CIFAR10, CIFAR100, and STL10) demonstrate the effectiveness of Corner CutMix, which consistently outperforms strong baselines such as CutOut and CutMix." @default.
- W4386597013 created "2023-09-12" @default.
- W4386597013 creator A5030374805 @default.
- W4386597013 creator A5052143634 @default.
- W4386597013 creator A5062125232 @default.
- W4386597013 creator A5077258552 @default.
- W4386597013 date "2023-10-08" @default.
- W4386597013 modified "2023-09-30" @default.
- W4386597013 title "Data Augmentation Using Corner CutMix and an Auxiliary Self-Supervised Loss" @default.
- W4386597013 cites W2194775991 @default.
- W4386597013 cites W2900595477 @default.
- W4386597013 cites W2949736877 @default.
- W4386597013 cites W2963045696 @default.
- W4386597013 cites W2992308087 @default.
- W4386597013 cites W2998508940 @default.
- W4386597013 cites W3035524453 @default.
- W4386597013 cites W3035682985 @default.
- W4386597013 doi "https://doi.org/10.1109/icip49359.2023.10222009" @default.
- W4386597013 hasPublicationYear "2023" @default.
- W4386597013 type Work @default.
- W4386597013 citedByCount "0" @default.
- W4386597013 crossrefType "proceedings-article" @default.
- W4386597013 hasAuthorship W4386597013A5030374805 @default.
- W4386597013 hasAuthorship W4386597013A5052143634 @default.
- W4386597013 hasAuthorship W4386597013A5062125232 @default.
- W4386597013 hasAuthorship W4386597013A5077258552 @default.
- W4386597013 hasBestOaLocation W43865970131 @default.
- W4386597013 hasConcept C105795698 @default.
- W4386597013 hasConcept C111919701 @default.
- W4386597013 hasConcept C115961682 @default.
- W4386597013 hasConcept C119857082 @default.
- W4386597013 hasConcept C126780896 @default.
- W4386597013 hasConcept C153180895 @default.
- W4386597013 hasConcept C154945302 @default.
- W4386597013 hasConcept C162324750 @default.
- W4386597013 hasConcept C17744445 @default.
- W4386597013 hasConcept C187736073 @default.
- W4386597013 hasConcept C194257627 @default.
- W4386597013 hasConcept C199539241 @default.
- W4386597013 hasConcept C22019652 @default.
- W4386597013 hasConcept C27158222 @default.
- W4386597013 hasConcept C2776257435 @default.
- W4386597013 hasConcept C2776359362 @default.
- W4386597013 hasConcept C2779960059 @default.
- W4386597013 hasConcept C2780451532 @default.
- W4386597013 hasConcept C31258907 @default.
- W4386597013 hasConcept C33923547 @default.
- W4386597013 hasConcept C41008148 @default.
- W4386597013 hasConcept C50644808 @default.
- W4386597013 hasConcept C75294576 @default.
- W4386597013 hasConcept C81363708 @default.
- W4386597013 hasConcept C94625758 @default.
- W4386597013 hasConceptScore W4386597013C105795698 @default.
- W4386597013 hasConceptScore W4386597013C111919701 @default.
- W4386597013 hasConceptScore W4386597013C115961682 @default.
- W4386597013 hasConceptScore W4386597013C119857082 @default.
- W4386597013 hasConceptScore W4386597013C126780896 @default.
- W4386597013 hasConceptScore W4386597013C153180895 @default.
- W4386597013 hasConceptScore W4386597013C154945302 @default.
- W4386597013 hasConceptScore W4386597013C162324750 @default.
- W4386597013 hasConceptScore W4386597013C17744445 @default.
- W4386597013 hasConceptScore W4386597013C187736073 @default.
- W4386597013 hasConceptScore W4386597013C194257627 @default.
- W4386597013 hasConceptScore W4386597013C199539241 @default.
- W4386597013 hasConceptScore W4386597013C22019652 @default.
- W4386597013 hasConceptScore W4386597013C27158222 @default.
- W4386597013 hasConceptScore W4386597013C2776257435 @default.
- W4386597013 hasConceptScore W4386597013C2776359362 @default.
- W4386597013 hasConceptScore W4386597013C2779960059 @default.
- W4386597013 hasConceptScore W4386597013C2780451532 @default.
- W4386597013 hasConceptScore W4386597013C31258907 @default.
- W4386597013 hasConceptScore W4386597013C33923547 @default.
- W4386597013 hasConceptScore W4386597013C41008148 @default.
- W4386597013 hasConceptScore W4386597013C50644808 @default.
- W4386597013 hasConceptScore W4386597013C75294576 @default.
- W4386597013 hasConceptScore W4386597013C81363708 @default.
- W4386597013 hasConceptScore W4386597013C94625758 @default.
- W4386597013 hasLocation W43865970131 @default.
- W4386597013 hasOpenAccess W4386597013 @default.
- W4386597013 hasPrimaryLocation W43865970131 @default.
- W4386597013 hasRelatedWork W2742991909 @default.
- W4386597013 hasRelatedWork W2767651786 @default.
- W4386597013 hasRelatedWork W2912288872 @default.
- W4386597013 hasRelatedWork W2989932438 @default.
- W4386597013 hasRelatedWork W3012393889 @default.
- W4386597013 hasRelatedWork W3081496756 @default.
- W4386597013 hasRelatedWork W3099765033 @default.
- W4386597013 hasRelatedWork W4210794429 @default.
- W4386597013 hasRelatedWork W4220996320 @default.
- W4386597013 hasRelatedWork W785854688 @default.
- W4386597013 isParatext "false" @default.
- W4386597013 isRetracted "false" @default.
- W4386597013 workType "article" @default.