Matches in SemOpenAlex for { <https://semopenalex.org/work/W4378221133> ?p ?o ?g. }
Showing items 1 to 42 of
42
with 100 items per page.
- W4378221133 abstract "Scene text detection aims to detect text regions in the complex background, and deep learning-based methods have been the mainstream. To obtain robust performance, deep learning-based methods are data-hungry, and one efficient technique to obtain sufficient data is data augmentation. However, current data augmentation for scene text detection is 1) changing the whole image, which ignores the instance-level diversity, or 2) generating synthetic data using generative models, which requires extra training data. In this paper, we propose a self-compositional data augmentation (SDA) for scene text detection. Our SDA generates new data by changing the original text regions of one image with four types of variations: translation scaling, rotation, and curving, and putting the changed text regions back into random places of the same image. In specific, our SDA is an instance-level augmentation, which could be combined with image-level augmentation; SDA requires no extra training data, which could be easily adopted in different methods. We conducted extensive experiments with three state-of-the-art scene text detection methods on two public datasets. Using our SDA improves all methods on all datasets, and the improvements demonstrate the effectiveness and generality of our SDA." @default.
- W4378221133 created "2023-05-26" @default.
- W4378221133 creator A5000732007 @default.
- W4378221133 creator A5067600725 @default.
- W4378221133 creator A5084780334 @default.
- W4378221133 date "2023-05-25" @default.
- W4378221133 modified "2023-10-18" @default.
- W4378221133 title "Self-compositional data augmentation for scene text detection" @default.
- W4378221133 doi "https://doi.org/10.1117/12.2679227" @default.
- W4378221133 hasPublicationYear "2023" @default.
- W4378221133 type Work @default.
- W4378221133 citedByCount "0" @default.
- W4378221133 crossrefType "proceedings-article" @default.
- W4378221133 hasAuthorship W4378221133A5000732007 @default.
- W4378221133 hasAuthorship W4378221133A5067600725 @default.
- W4378221133 hasAuthorship W4378221133A5084780334 @default.
- W4378221133 hasConcept C121684516 @default.
- W4378221133 hasConcept C154945302 @default.
- W4378221133 hasConcept C204321447 @default.
- W4378221133 hasConcept C31972630 @default.
- W4378221133 hasConcept C41008148 @default.
- W4378221133 hasConceptScore W4378221133C121684516 @default.
- W4378221133 hasConceptScore W4378221133C154945302 @default.
- W4378221133 hasConceptScore W4378221133C204321447 @default.
- W4378221133 hasConceptScore W4378221133C31972630 @default.
- W4378221133 hasConceptScore W4378221133C41008148 @default.
- W4378221133 hasLocation W43782211331 @default.
- W4378221133 hasOpenAccess W4378221133 @default.
- W4378221133 hasPrimaryLocation W43782211331 @default.
- W4378221133 hasRelatedWork W1891287906 @default.
- W4378221133 hasRelatedWork W1969923398 @default.
- W4378221133 hasRelatedWork W2036807459 @default.
- W4378221133 hasRelatedWork W2058170566 @default.
- W4378221133 hasRelatedWork W2166024367 @default.
- W4378221133 hasRelatedWork W2166044122 @default.
- W4378221133 hasRelatedWork W2229312674 @default.
- W4378221133 hasRelatedWork W2755342338 @default.
- W4378221133 hasRelatedWork W2772917594 @default.
- W4378221133 hasRelatedWork W3116076068 @default.
- W4378221133 isParatext "false" @default.
- W4378221133 isRetracted "false" @default.
- W4378221133 workType "article" @default.