Matches in SemOpenAlex for { <https://semopenalex.org/work/W3088045114> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W3088045114 abstract "Deep learning methods for automatic visual inspection can help detect potential anomalies in real-time operations and reduce the burden of full-length video inspection analysis. Such learning-based methods require, however, huge amounts of diverse labeled examples to be trained on. The lack of variance in examples from less frequent entities (eg. anomalies) makes it difficult for deep models to generalize such tasks, leading to poor performance. We then propose to use multitask learning to improve the generalization capacity, thus the performance of deep classification models at the recognition of less frequent entities in the underwater domain. In particular, we exploit hierarchical relationships between classification tasks to wisely choose which tasks to combine. For the imbalanced task labet, our approach is shown to improve multitask performance from 0.74 to 0.89, in terms of average precision. Efficiency is also improved, with inference time being reduced by almost fifty percent, thus making it significantly cheaper to use visual inspection models in realtime operations." @default.
- W3088045114 created "2020-10-01" @default.
- W3088045114 creator A5000868838 @default.
- W3088045114 creator A5023367656 @default.
- W3088045114 creator A5029044972 @default.
- W3088045114 creator A5061171324 @default.
- W3088045114 date "2020-09-01" @default.
- W3088045114 modified "2023-09-25" @default.
- W3088045114 title "Hierarchical Multitask Learning for Improved Underwater Recognition on Imbalanced Tasks" @default.
- W3088045114 cites W167016754 @default.
- W3088045114 cites W1980501707 @default.
- W3088045114 cites W2108598243 @default.
- W3088045114 cites W2186054958 @default.
- W3088045114 cites W2187558063 @default.
- W3088045114 cites W2407277018 @default.
- W3088045114 cites W2590925815 @default.
- W3088045114 cites W2624275270 @default.
- W3088045114 cites W2794869810 @default.
- W3088045114 cites W2962743139 @default.
- W3088045114 cites W2963524571 @default.
- W3088045114 cites W2963704251 @default.
- W3088045114 cites W2963877604 @default.
- W3088045114 cites W2970286909 @default.
- W3088045114 cites W2998175747 @default.
- W3088045114 cites W2962764337 @default.
- W3088045114 doi "https://doi.org/10.1109/icist49303.2020.9202063" @default.
- W3088045114 hasPublicationYear "2020" @default.
- W3088045114 type Work @default.
- W3088045114 sameAs 3088045114 @default.
- W3088045114 citedByCount "0" @default.
- W3088045114 crossrefType "proceedings-article" @default.
- W3088045114 hasAuthorship W3088045114A5000868838 @default.
- W3088045114 hasAuthorship W3088045114A5023367656 @default.
- W3088045114 hasAuthorship W3088045114A5029044972 @default.
- W3088045114 hasAuthorship W3088045114A5061171324 @default.
- W3088045114 hasConcept C108583219 @default.
- W3088045114 hasConcept C111368507 @default.
- W3088045114 hasConcept C119857082 @default.
- W3088045114 hasConcept C121955636 @default.
- W3088045114 hasConcept C127313418 @default.
- W3088045114 hasConcept C127413603 @default.
- W3088045114 hasConcept C134306372 @default.
- W3088045114 hasConcept C144133560 @default.
- W3088045114 hasConcept C154945302 @default.
- W3088045114 hasConcept C165696696 @default.
- W3088045114 hasConcept C175154964 @default.
- W3088045114 hasConcept C177148314 @default.
- W3088045114 hasConcept C196083921 @default.
- W3088045114 hasConcept C201995342 @default.
- W3088045114 hasConcept C2776214188 @default.
- W3088045114 hasConcept C2780451532 @default.
- W3088045114 hasConcept C28006648 @default.
- W3088045114 hasConcept C33923547 @default.
- W3088045114 hasConcept C36503486 @default.
- W3088045114 hasConcept C38652104 @default.
- W3088045114 hasConcept C41008148 @default.
- W3088045114 hasConcept C98083399 @default.
- W3088045114 hasConceptScore W3088045114C108583219 @default.
- W3088045114 hasConceptScore W3088045114C111368507 @default.
- W3088045114 hasConceptScore W3088045114C119857082 @default.
- W3088045114 hasConceptScore W3088045114C121955636 @default.
- W3088045114 hasConceptScore W3088045114C127313418 @default.
- W3088045114 hasConceptScore W3088045114C127413603 @default.
- W3088045114 hasConceptScore W3088045114C134306372 @default.
- W3088045114 hasConceptScore W3088045114C144133560 @default.
- W3088045114 hasConceptScore W3088045114C154945302 @default.
- W3088045114 hasConceptScore W3088045114C165696696 @default.
- W3088045114 hasConceptScore W3088045114C175154964 @default.
- W3088045114 hasConceptScore W3088045114C177148314 @default.
- W3088045114 hasConceptScore W3088045114C196083921 @default.
- W3088045114 hasConceptScore W3088045114C201995342 @default.
- W3088045114 hasConceptScore W3088045114C2776214188 @default.
- W3088045114 hasConceptScore W3088045114C2780451532 @default.
- W3088045114 hasConceptScore W3088045114C28006648 @default.
- W3088045114 hasConceptScore W3088045114C33923547 @default.
- W3088045114 hasConceptScore W3088045114C36503486 @default.
- W3088045114 hasConceptScore W3088045114C38652104 @default.
- W3088045114 hasConceptScore W3088045114C41008148 @default.
- W3088045114 hasConceptScore W3088045114C98083399 @default.
- W3088045114 hasLocation W30880451141 @default.
- W3088045114 hasOpenAccess W3088045114 @default.
- W3088045114 hasPrimaryLocation W30880451141 @default.
- W3088045114 hasRelatedWork W10944326 @default.
- W3088045114 hasRelatedWork W11991885 @default.
- W3088045114 hasRelatedWork W12016155 @default.
- W3088045114 hasRelatedWork W13538861 @default.
- W3088045114 hasRelatedWork W14789944 @default.
- W3088045114 hasRelatedWork W354571 @default.
- W3088045114 hasRelatedWork W3569651 @default.
- W3088045114 hasRelatedWork W7594470 @default.
- W3088045114 hasRelatedWork W868042 @default.
- W3088045114 hasRelatedWork W722847 @default.
- W3088045114 isParatext "false" @default.
- W3088045114 isRetracted "false" @default.
- W3088045114 magId "3088045114" @default.
- W3088045114 workType "article" @default.