Matches in SemOpenAlex for { <https://semopenalex.org/work/W2977807402> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W2977807402 abstract "This paper proposes a novel attention mechanism and a fancy loss function for scene text detectors. Specifically, the attention mechanism can effectively identify the text regions by learning an attention mask automatically. The fine-grained attention mask is directly incorporated into the convolutional feature maps of a neural network to produce graininess-aware feature maps, which essentially obstruct the background inference and especially emphasize the text regions. Therefore, our graininess-aware feature maps concentrate on text regions, in especial those of exceedingly small size. Additionally, to address the extreme text-background class imbalance during training, we also propose a newfangled loss function, named Focal Negative Loss (FNL). The proposed loss function is able to down-weight the loss assigned to easy negative samples. Consequently, the proposed FNL can make training focused on hard negative samples. To evaluate the effectiveness of our text attention module and FNL, we integrate them into the efficient and accurate scene text detector (EAST). The comprehensive experimental results demonstrate that our text attention module and FNL can increase the performance of EAST by F-score of 3.98% on ICDAR2015 dataset and 1.87% on MSRA-TD500 dataset." @default.
- W2977807402 created "2019-10-10" @default.
- W2977807402 creator A5023759358 @default.
- W2977807402 creator A5075782732 @default.
- W2977807402 date "2019-07-01" @default.
- W2977807402 modified "2023-09-25" @default.
- W2977807402 title "Text Attention and Focal Negative Loss for Scene Text Detection" @default.
- W2977807402 cites W117491841 @default.
- W2977807402 cites W1922126009 @default.
- W2977807402 cites W1988461287 @default.
- W2977807402 cites W2006653496 @default.
- W2977807402 cites W2008806374 @default.
- W2977807402 cites W2019478948 @default.
- W2977807402 cites W2061802763 @default.
- W2977807402 cites W2102605133 @default.
- W2977807402 cites W2144554289 @default.
- W2977807402 cites W2194775991 @default.
- W2977807402 cites W2216125271 @default.
- W2977807402 cites W2339589954 @default.
- W2977807402 cites W2343052201 @default.
- W2977807402 cites W2468724597 @default.
- W2977807402 cites W2472159136 @default.
- W2977807402 cites W2519818067 @default.
- W2977807402 cites W2604243686 @default.
- W2977807402 cites W2604735854 @default.
- W2977807402 cites W2605076167 @default.
- W2977807402 cites W2605982830 @default.
- W2977807402 cites W2831607544 @default.
- W2977807402 cites W2895077992 @default.
- W2977807402 cites W2962810613 @default.
- W2977807402 cites W2963037989 @default.
- W2977807402 cites W2963150697 @default.
- W2977807402 cites W2963187132 @default.
- W2977807402 cites W2963195262 @default.
- W2977807402 cites W2963351448 @default.
- W2977807402 cites W2963840241 @default.
- W2977807402 cites W2963977642 @default.
- W2977807402 cites W2964018263 @default.
- W2977807402 cites W3098090606 @default.
- W2977807402 cites W3106228955 @default.
- W2977807402 doi "https://doi.org/10.1109/ijcnn.2019.8851959" @default.
- W2977807402 hasPublicationYear "2019" @default.
- W2977807402 type Work @default.
- W2977807402 sameAs 2977807402 @default.
- W2977807402 citedByCount "2" @default.
- W2977807402 countsByYear W29778074022021 @default.
- W2977807402 countsByYear W29778074022022 @default.
- W2977807402 crossrefType "proceedings-article" @default.
- W2977807402 hasAuthorship W2977807402A5023759358 @default.
- W2977807402 hasAuthorship W2977807402A5075782732 @default.
- W2977807402 hasConcept C108583219 @default.
- W2977807402 hasConcept C138885662 @default.
- W2977807402 hasConcept C14036430 @default.
- W2977807402 hasConcept C153180895 @default.
- W2977807402 hasConcept C154945302 @default.
- W2977807402 hasConcept C2776214188 @default.
- W2977807402 hasConcept C2776401178 @default.
- W2977807402 hasConcept C2777212361 @default.
- W2977807402 hasConcept C41008148 @default.
- W2977807402 hasConcept C41895202 @default.
- W2977807402 hasConcept C78458016 @default.
- W2977807402 hasConcept C81363708 @default.
- W2977807402 hasConcept C86803240 @default.
- W2977807402 hasConceptScore W2977807402C108583219 @default.
- W2977807402 hasConceptScore W2977807402C138885662 @default.
- W2977807402 hasConceptScore W2977807402C14036430 @default.
- W2977807402 hasConceptScore W2977807402C153180895 @default.
- W2977807402 hasConceptScore W2977807402C154945302 @default.
- W2977807402 hasConceptScore W2977807402C2776214188 @default.
- W2977807402 hasConceptScore W2977807402C2776401178 @default.
- W2977807402 hasConceptScore W2977807402C2777212361 @default.
- W2977807402 hasConceptScore W2977807402C41008148 @default.
- W2977807402 hasConceptScore W2977807402C41895202 @default.
- W2977807402 hasConceptScore W2977807402C78458016 @default.
- W2977807402 hasConceptScore W2977807402C81363708 @default.
- W2977807402 hasConceptScore W2977807402C86803240 @default.
- W2977807402 hasLocation W29778074021 @default.
- W2977807402 hasOpenAccess W2977807402 @default.
- W2977807402 hasPrimaryLocation W29778074021 @default.
- W2977807402 hasRelatedWork W2731899572 @default.
- W2977807402 hasRelatedWork W2999805992 @default.
- W2977807402 hasRelatedWork W3011074480 @default.
- W2977807402 hasRelatedWork W3116150086 @default.
- W2977807402 hasRelatedWork W3133861977 @default.
- W2977807402 hasRelatedWork W3192840557 @default.
- W2977807402 hasRelatedWork W4200173597 @default.
- W2977807402 hasRelatedWork W4291897433 @default.
- W2977807402 hasRelatedWork W4312417841 @default.
- W2977807402 hasRelatedWork W4321369474 @default.
- W2977807402 isParatext "false" @default.
- W2977807402 isRetracted "false" @default.
- W2977807402 magId "2977807402" @default.
- W2977807402 workType "article" @default.