Matches in SemOpenAlex for { <https://semopenalex.org/work/W3148786143> ?p ?o ?g. }
- W3148786143 endingPage "167" @default.
- W3148786143 startingPage "152" @default.
- W3148786143 abstract "Existing methods based on convolutional neural networks (CNNs) have progressed significantly in the salient object detection task. However, in some cases, the accurate detection of intact objects and maintenance of their original detailed structures, such as boundaries, was not possible. In this paper, we propose an Enhancing Region and Boundary Awareness Network (ERBANet), equipped with attentional feature enhancement (AFE) modules to improve the detection performance. The AFE modules act on high-level and low-level features to generate corresponding attentional features. High-level attentional features are used to highlight entire salient objects, while low-level attentional features help retain their boundaries. The proposed ERBANet aims to effectively aggregate high-level and low-level attentional features, fully utilizing their respective advantages. Furthermore, we propose a novel boundary maintenance loss (BML) for learning to preserve the original boundaries of salient objects. Meanwhile, dice loss is combined for learning to enhance the integrity of salient regions. The experimental results on five benchmark datasets demonstrate that our proposed method outperforms recent state-of-the-art methods and achieves better performance." @default.
- W3148786143 created "2021-04-13" @default.
- W3148786143 creator A5003702543 @default.
- W3148786143 creator A5072825282 @default.
- W3148786143 date "2021-08-01" @default.
- W3148786143 modified "2023-10-17" @default.
- W3148786143 title "ERBANet: Enhancing region and boundary awareness for salient object detection" @default.
- W3148786143 cites W1978479866 @default.
- W3148786143 cites W2027850463 @default.
- W3148786143 cites W2098702446 @default.
- W3148786143 cites W2128272608 @default.
- W3148786143 cites W2160398355 @default.
- W3148786143 cites W2519623608 @default.
- W3148786143 cites W2520274358 @default.
- W3148786143 cites W2585592883 @default.
- W3148786143 cites W2588503511 @default.
- W3148786143 cites W2592997798 @default.
- W3148786143 cites W2757028014 @default.
- W3148786143 cites W2807746031 @default.
- W3148786143 cites W2884585870 @default.
- W3148786143 cites W2904454836 @default.
- W3148786143 cites W2905825933 @default.
- W3148786143 cites W2921653116 @default.
- W3148786143 cites W2955060956 @default.
- W3148786143 cites W2963393688 @default.
- W3148786143 cites W2963422510 @default.
- W3148786143 cites W2963450895 @default.
- W3148786143 cites W2963647456 @default.
- W3148786143 cites W2963868681 @default.
- W3148786143 cites W2963893037 @default.
- W3148786143 cites W2964309882 @default.
- W3148786143 cites W2990278031 @default.
- W3148786143 cites W2995743241 @default.
- W3148786143 cites W2999674884 @default.
- W3148786143 cites W3000281486 @default.
- W3148786143 cites W3017990003 @default.
- W3148786143 cites W3034326710 @default.
- W3148786143 cites W3034879371 @default.
- W3148786143 cites W3087221416 @default.
- W3148786143 cites W3098389804 @default.
- W3148786143 cites W3104943862 @default.
- W3148786143 cites W3104979525 @default.
- W3148786143 cites W4239147634 @default.
- W3148786143 cites W4250482878 @default.
- W3148786143 doi "https://doi.org/10.1016/j.neucom.2021.03.094" @default.
- W3148786143 hasPublicationYear "2021" @default.
- W3148786143 type Work @default.
- W3148786143 sameAs 3148786143 @default.
- W3148786143 citedByCount "13" @default.
- W3148786143 countsByYear W31487861432021 @default.
- W3148786143 countsByYear W31487861432022 @default.
- W3148786143 countsByYear W31487861432023 @default.
- W3148786143 crossrefType "journal-article" @default.
- W3148786143 hasAuthorship W3148786143A5003702543 @default.
- W3148786143 hasAuthorship W3148786143A5072825282 @default.
- W3148786143 hasConcept C127413603 @default.
- W3148786143 hasConcept C13280743 @default.
- W3148786143 hasConcept C134306372 @default.
- W3148786143 hasConcept C138885662 @default.
- W3148786143 hasConcept C153180895 @default.
- W3148786143 hasConcept C154945302 @default.
- W3148786143 hasConcept C185798385 @default.
- W3148786143 hasConcept C201995342 @default.
- W3148786143 hasConcept C205649164 @default.
- W3148786143 hasConcept C2776151529 @default.
- W3148786143 hasConcept C2776401178 @default.
- W3148786143 hasConcept C2780451532 @default.
- W3148786143 hasConcept C2780719617 @default.
- W3148786143 hasConcept C33923547 @default.
- W3148786143 hasConcept C41008148 @default.
- W3148786143 hasConcept C41895202 @default.
- W3148786143 hasConcept C59404180 @default.
- W3148786143 hasConcept C62354387 @default.
- W3148786143 hasConcept C81363708 @default.
- W3148786143 hasConceptScore W3148786143C127413603 @default.
- W3148786143 hasConceptScore W3148786143C13280743 @default.
- W3148786143 hasConceptScore W3148786143C134306372 @default.
- W3148786143 hasConceptScore W3148786143C138885662 @default.
- W3148786143 hasConceptScore W3148786143C153180895 @default.
- W3148786143 hasConceptScore W3148786143C154945302 @default.
- W3148786143 hasConceptScore W3148786143C185798385 @default.
- W3148786143 hasConceptScore W3148786143C201995342 @default.
- W3148786143 hasConceptScore W3148786143C205649164 @default.
- W3148786143 hasConceptScore W3148786143C2776151529 @default.
- W3148786143 hasConceptScore W3148786143C2776401178 @default.
- W3148786143 hasConceptScore W3148786143C2780451532 @default.
- W3148786143 hasConceptScore W3148786143C2780719617 @default.
- W3148786143 hasConceptScore W3148786143C33923547 @default.
- W3148786143 hasConceptScore W3148786143C41008148 @default.
- W3148786143 hasConceptScore W3148786143C41895202 @default.
- W3148786143 hasConceptScore W3148786143C59404180 @default.
- W3148786143 hasConceptScore W3148786143C62354387 @default.
- W3148786143 hasConceptScore W3148786143C81363708 @default.
- W3148786143 hasLocation W31487861431 @default.
- W3148786143 hasOpenAccess W3148786143 @default.
- W3148786143 hasPrimaryLocation W31487861431 @default.
- W3148786143 hasRelatedWork W2059299633 @default.
- W3148786143 hasRelatedWork W2296151615 @default.