Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386134471> ?p ?o ?g. }
- W4386134471 endingPage "126742" @default.
- W4386134471 startingPage "126742" @default.
- W4386134471 abstract "Existing Deep-Learning-based contour detectors suffer from the issues of the sharpness and the correctness of their predictions, which often need offline post-processing to sharpen the results as well as improve model performance. In this work, we present a novel method that can learn to refine object contour in training and directly output crisp object boundaries in inference. To this end, we first introduce a keypoint-focal loss that draws point-based attention to the isolated contour annotations. It allows an edge detector to jointly optimize the appearance thickness and the localization accuracy of predictions in the training procedure. Moreover, we present a regularization loss to further improve the performance of an edge detector. Lastly, we present the Contour Transformer model for precisely localizing object boundaries in images. We repurpose and integrate a Transformer-style hyper module into an encoder–decoder network, effectively aggregating global contextual information on high-level features and significantly enhancing the discriminative power for classifying foreground/background pixels. We train and test our Attention and Contour Transformer detector (ACTD) on four widely adopted datasets, i.e., BSDS500, NYUD, Multi-Cue, and RoadNet. The proposed method achieves an ODS F-score of 0.826 on BSDS500 and an ODS F-score of 0.783 on NYUD, outperforming previous top detectors." @default.
- W4386134471 created "2023-08-25" @default.
- W4386134471 creator A5004212956 @default.
- W4386134471 creator A5024138459 @default.
- W4386134471 creator A5034394648 @default.
- W4386134471 creator A5039152554 @default.
- W4386134471 date "2023-11-01" @default.
- W4386134471 modified "2023-10-17" @default.
- W4386134471 title "Learning to refine object boundaries" @default.
- W4386134471 cites W1976047850 @default.
- W4386134471 cites W2016396776 @default.
- W4386134471 cites W2071356085 @default.
- W4386134471 cites W2110158442 @default.
- W4386134471 cites W2145023731 @default.
- W4386134471 cites W2150134853 @default.
- W4386134471 cites W2151103935 @default.
- W4386134471 cites W2152210568 @default.
- W4386134471 cites W2289871033 @default.
- W4386134471 cites W2560622558 @default.
- W4386134471 cites W2566356915 @default.
- W4386134471 cites W2607333215 @default.
- W4386134471 cites W2769933874 @default.
- W4386134471 cites W2898504016 @default.
- W4386134471 cites W3010722397 @default.
- W4386134471 cites W3037327307 @default.
- W4386134471 cites W4226442776 @default.
- W4386134471 cites W4283781662 @default.
- W4386134471 cites W4283811313 @default.
- W4386134471 cites W4307593471 @default.
- W4386134471 cites W4312055535 @default.
- W4386134471 cites W4312058481 @default.
- W4386134471 doi "https://doi.org/10.1016/j.neucom.2023.126742" @default.
- W4386134471 hasPublicationYear "2023" @default.
- W4386134471 type Work @default.
- W4386134471 citedByCount "0" @default.
- W4386134471 crossrefType "journal-article" @default.
- W4386134471 hasAuthorship W4386134471A5004212956 @default.
- W4386134471 hasAuthorship W4386134471A5024138459 @default.
- W4386134471 hasAuthorship W4386134471A5034394648 @default.
- W4386134471 hasAuthorship W4386134471A5039152554 @default.
- W4386134471 hasConcept C111919701 @default.
- W4386134471 hasConcept C11413529 @default.
- W4386134471 hasConcept C118505674 @default.
- W4386134471 hasConcept C121332964 @default.
- W4386134471 hasConcept C138236772 @default.
- W4386134471 hasConcept C153180895 @default.
- W4386134471 hasConcept C154945302 @default.
- W4386134471 hasConcept C160633673 @default.
- W4386134471 hasConcept C165801399 @default.
- W4386134471 hasConcept C2776135515 @default.
- W4386134471 hasConcept C2776151529 @default.
- W4386134471 hasConcept C2776214188 @default.
- W4386134471 hasConcept C31972630 @default.
- W4386134471 hasConcept C41008148 @default.
- W4386134471 hasConcept C55439883 @default.
- W4386134471 hasConcept C62520636 @default.
- W4386134471 hasConcept C66322947 @default.
- W4386134471 hasConcept C76155785 @default.
- W4386134471 hasConcept C79974875 @default.
- W4386134471 hasConcept C94915269 @default.
- W4386134471 hasConcept C97931131 @default.
- W4386134471 hasConceptScore W4386134471C111919701 @default.
- W4386134471 hasConceptScore W4386134471C11413529 @default.
- W4386134471 hasConceptScore W4386134471C118505674 @default.
- W4386134471 hasConceptScore W4386134471C121332964 @default.
- W4386134471 hasConceptScore W4386134471C138236772 @default.
- W4386134471 hasConceptScore W4386134471C153180895 @default.
- W4386134471 hasConceptScore W4386134471C154945302 @default.
- W4386134471 hasConceptScore W4386134471C160633673 @default.
- W4386134471 hasConceptScore W4386134471C165801399 @default.
- W4386134471 hasConceptScore W4386134471C2776135515 @default.
- W4386134471 hasConceptScore W4386134471C2776151529 @default.
- W4386134471 hasConceptScore W4386134471C2776214188 @default.
- W4386134471 hasConceptScore W4386134471C31972630 @default.
- W4386134471 hasConceptScore W4386134471C41008148 @default.
- W4386134471 hasConceptScore W4386134471C55439883 @default.
- W4386134471 hasConceptScore W4386134471C62520636 @default.
- W4386134471 hasConceptScore W4386134471C66322947 @default.
- W4386134471 hasConceptScore W4386134471C76155785 @default.
- W4386134471 hasConceptScore W4386134471C79974875 @default.
- W4386134471 hasConceptScore W4386134471C94915269 @default.
- W4386134471 hasConceptScore W4386134471C97931131 @default.
- W4386134471 hasFunder F4320321001 @default.
- W4386134471 hasFunder F4320335581 @default.
- W4386134471 hasLocation W43861344711 @default.
- W4386134471 hasOpenAccess W4386134471 @default.
- W4386134471 hasPrimaryLocation W43861344711 @default.
- W4386134471 hasRelatedWork W2024160000 @default.
- W4386134471 hasRelatedWork W2061273563 @default.
- W4386134471 hasRelatedWork W2136485282 @default.
- W4386134471 hasRelatedWork W2285052147 @default.
- W4386134471 hasRelatedWork W2546871836 @default.
- W4386134471 hasRelatedWork W2729514902 @default.
- W4386134471 hasRelatedWork W2773500201 @default.
- W4386134471 hasRelatedWork W4287995534 @default.
- W4386134471 hasRelatedWork W4312857205 @default.
- W4386134471 hasRelatedWork W2073139667 @default.
- W4386134471 hasVolume "557" @default.