Matches in SemOpenAlex for { <https://semopenalex.org/work/W4304014014> ?p ?o ?g. }
Showing items 1 to 82 of
82
with 100 items per page.
- W4304014014 abstract "Image Transformer has recently achieved significant progress for natural image understanding, either using supervised (ViT, DeiT, etc.) or self-supervised (BEiT, MAE, etc.) pre-training techniques. In this paper, we propose DiT, a self-supervised pre-trained Document Image Transformer model using large-scale unlabeled text images for Document AI tasks, which is essential since no supervised counterparts ever exist due to the lack of human-labeled document images. We leverage DiT as the backbone network in a variety of vision-based Document AI tasks, including document image classification, document layout analysis, table detection as well as text detection for OCR. Experiment results have illustrated that the self-supervised pre-trained DiT model achieves new state-of-the-art results on these downstream tasks, e.g. document image classification (91.11 - 92.69), document layout analysis (91.0 - 94.9), table detection (94.23 - 96.55) and text detection for OCR (93.07 - 94.29). The code and pre-trained models are publicly available at https://aka.ms/msdit." @default.
- W4304014014 created "2022-10-10" @default.
- W4304014014 creator A5014662947 @default.
- W4304014014 creator A5026963407 @default.
- W4304014014 creator A5031963349 @default.
- W4304014014 creator A5038013079 @default.
- W4304014014 creator A5059884905 @default.
- W4304014014 creator A5077320577 @default.
- W4304014014 date "2022-10-10" @default.
- W4304014014 modified "2023-10-16" @default.
- W4304014014 title "DiT: Self-supervised Pre-training for Document Image Transformer" @default.
- W4304014014 cites W2605976347 @default.
- W4304014014 cites W2965085721 @default.
- W4304014014 cites W3003711898 @default.
- W4304014014 cites W3096609285 @default.
- W4304014014 cites W3107064625 @default.
- W4304014014 cites W3113753692 @default.
- W4304014014 cites W3145450063 @default.
- W4304014014 cites W3159481202 @default.
- W4304014014 cites W3202839357 @default.
- W4304014014 cites W4304013646 @default.
- W4304014014 doi "https://doi.org/10.1145/3503161.3547911" @default.
- W4304014014 hasPublicationYear "2022" @default.
- W4304014014 type Work @default.
- W4304014014 citedByCount "16" @default.
- W4304014014 countsByYear W43040140142022 @default.
- W4304014014 countsByYear W43040140142023 @default.
- W4304014014 crossrefType "proceedings-article" @default.
- W4304014014 hasAuthorship W4304014014A5014662947 @default.
- W4304014014 hasAuthorship W4304014014A5026963407 @default.
- W4304014014 hasAuthorship W4304014014A5031963349 @default.
- W4304014014 hasAuthorship W4304014014A5038013079 @default.
- W4304014014 hasAuthorship W4304014014A5059884905 @default.
- W4304014014 hasAuthorship W4304014014A5077320577 @default.
- W4304014014 hasBestOaLocation W43040140142 @default.
- W4304014014 hasConcept C115961682 @default.
- W4304014014 hasConcept C119599485 @default.
- W4304014014 hasConcept C119857082 @default.
- W4304014014 hasConcept C127413603 @default.
- W4304014014 hasConcept C136389625 @default.
- W4304014014 hasConcept C153083717 @default.
- W4304014014 hasConcept C153180895 @default.
- W4304014014 hasConcept C154945302 @default.
- W4304014014 hasConcept C165801399 @default.
- W4304014014 hasConcept C204321447 @default.
- W4304014014 hasConcept C23123220 @default.
- W4304014014 hasConcept C41008148 @default.
- W4304014014 hasConcept C50644808 @default.
- W4304014014 hasConcept C66322947 @default.
- W4304014014 hasConcept C72773152 @default.
- W4304014014 hasConceptScore W4304014014C115961682 @default.
- W4304014014 hasConceptScore W4304014014C119599485 @default.
- W4304014014 hasConceptScore W4304014014C119857082 @default.
- W4304014014 hasConceptScore W4304014014C127413603 @default.
- W4304014014 hasConceptScore W4304014014C136389625 @default.
- W4304014014 hasConceptScore W4304014014C153083717 @default.
- W4304014014 hasConceptScore W4304014014C153180895 @default.
- W4304014014 hasConceptScore W4304014014C154945302 @default.
- W4304014014 hasConceptScore W4304014014C165801399 @default.
- W4304014014 hasConceptScore W4304014014C204321447 @default.
- W4304014014 hasConceptScore W4304014014C23123220 @default.
- W4304014014 hasConceptScore W4304014014C41008148 @default.
- W4304014014 hasConceptScore W4304014014C50644808 @default.
- W4304014014 hasConceptScore W4304014014C66322947 @default.
- W4304014014 hasConceptScore W4304014014C72773152 @default.
- W4304014014 hasLocation W43040140141 @default.
- W4304014014 hasLocation W43040140142 @default.
- W4304014014 hasOpenAccess W4304014014 @default.
- W4304014014 hasPrimaryLocation W43040140141 @default.
- W4304014014 hasRelatedWork W2130283001 @default.
- W4304014014 hasRelatedWork W2158269427 @default.
- W4304014014 hasRelatedWork W2275805942 @default.
- W4304014014 hasRelatedWork W2355048207 @default.
- W4304014014 hasRelatedWork W2750422482 @default.
- W4304014014 hasRelatedWork W2787993192 @default.
- W4304014014 hasRelatedWork W2847365777 @default.
- W4304014014 hasRelatedWork W3033859939 @default.
- W4304014014 hasRelatedWork W3126051647 @default.
- W4304014014 hasRelatedWork W4381280689 @default.
- W4304014014 isParatext "false" @default.
- W4304014014 isRetracted "false" @default.
- W4304014014 workType "article" @default.