Matches in SemOpenAlex for { <https://semopenalex.org/work/W3100877062> ?p ?o ?g. }
- W3100877062 endingPage "1914" @default.
- W3100877062 startingPage "1902" @default.
- W3100877062 abstract "The performance of video saliency estimation techniques has achieved significant advances along with the rapid development of Convolutional Neural Networks (CNNs). However, devices like cameras and drones may have limited computational capability and storage space so that the direct deployment of complex deep saliency models becomes infeasible. To address this problem, this paper proposes a dynamic saliency estimation approach for aerial videos via spatiotemporal knowledge distillation. In this approach, five components are involved, including two teachers, two students and the desired spatiotemporal model. The knowledge of spatial and temporal saliency is first separately transferred from the two complex and redundant teachers to their simple and compact students, while the input scenes are also degraded from high-resolution to low-resolution to remove the probable data redundancy so as to greatly speed up the feature extraction process. After that, the desired spatiotemporal model is further trained by distilling and encoding the spatial and temporal saliency knowledge of two students into a unified network. In this manner, the inter-model redundancy can be removed for the effective estimation of dynamic saliency on aerial videos. Experimental results show that the proposed approach is comparable to 11 state-of-the-art models in estimating visual saliency on aerial videos, while its speed reaches up to 28,738 FPS and 1,490.5 FPS on the GPU and CPU platforms, respectively." @default.
- W3100877062 created "2020-11-23" @default.
- W3100877062 creator A5033254559 @default.
- W3100877062 creator A5059344854 @default.
- W3100877062 creator A5064644462 @default.
- W3100877062 creator A5084763603 @default.
- W3100877062 date "2020-01-01" @default.
- W3100877062 modified "2023-10-16" @default.
- W3100877062 title "Spatiotemporal Knowledge Distillation for Efficient Estimation of Aerial Video Saliency" @default.
- W3100877062 cites W1902041153 @default.
- W3100877062 cites W1959094031 @default.
- W3100877062 cites W1970342316 @default.
- W3100877062 cites W1985202143 @default.
- W3100877062 cites W2027253213 @default.
- W3100877062 cites W2033859430 @default.
- W3100877062 cites W2034436892 @default.
- W3100877062 cites W2039298799 @default.
- W3100877062 cites W2041719651 @default.
- W3100877062 cites W2047275730 @default.
- W3100877062 cites W2051111292 @default.
- W3100877062 cites W2078903912 @default.
- W3100877062 cites W2081239020 @default.
- W3100877062 cites W2095816583 @default.
- W3100877062 cites W2098702446 @default.
- W3100877062 cites W2104636679 @default.
- W3100877062 cites W2108481495 @default.
- W3100877062 cites W2108598243 @default.
- W3100877062 cites W2110019070 @default.
- W3100877062 cites W2113808907 @default.
- W3100877062 cites W2115717322 @default.
- W3100877062 cites W2128272608 @default.
- W3100877062 cites W2130184867 @default.
- W3100877062 cites W2131297486 @default.
- W3100877062 cites W2147347517 @default.
- W3100877062 cites W2148383759 @default.
- W3100877062 cites W2151096523 @default.
- W3100877062 cites W2152233525 @default.
- W3100877062 cites W2158983298 @default.
- W3100877062 cites W2207629083 @default.
- W3100877062 cites W2212216676 @default.
- W3100877062 cites W2288514685 @default.
- W3100877062 cites W2294370754 @default.
- W3100877062 cites W2295252581 @default.
- W3100877062 cites W2342927252 @default.
- W3100877062 cites W2378845821 @default.
- W3100877062 cites W2528092473 @default.
- W3100877062 cites W2533058588 @default.
- W3100877062 cites W2562902982 @default.
- W3100877062 cites W2585592883 @default.
- W3100877062 cites W2600389231 @default.
- W3100877062 cites W2620998106 @default.
- W3100877062 cites W2718471533 @default.
- W3100877062 cites W2733584164 @default.
- W3100877062 cites W2739879705 @default.
- W3100877062 cites W2740652190 @default.
- W3100877062 cites W2757028014 @default.
- W3100877062 cites W2782834210 @default.
- W3100877062 cites W2810970355 @default.
- W3100877062 cites W2895340898 @default.
- W3100877062 cites W2962965915 @default.
- W3100877062 cites W2963503775 @default.
- W3100877062 cites W2963581854 @default.
- W3100877062 cites W2963635628 @default.
- W3100877062 cites W3099561715 @default.
- W3100877062 cites W3101787898 @default.
- W3100877062 cites W3122238731 @default.
- W3100877062 doi "https://doi.org/10.1109/tip.2019.2946102" @default.
- W3100877062 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/31613767" @default.
- W3100877062 hasPublicationYear "2020" @default.
- W3100877062 type Work @default.
- W3100877062 sameAs 3100877062 @default.
- W3100877062 citedByCount "23" @default.
- W3100877062 countsByYear W31008770622019 @default.
- W3100877062 countsByYear W31008770622020 @default.
- W3100877062 countsByYear W31008770622021 @default.
- W3100877062 countsByYear W31008770622022 @default.
- W3100877062 countsByYear W31008770622023 @default.
- W3100877062 crossrefType "journal-article" @default.
- W3100877062 hasAuthorship W3100877062A5033254559 @default.
- W3100877062 hasAuthorship W3100877062A5059344854 @default.
- W3100877062 hasAuthorship W3100877062A5064644462 @default.
- W3100877062 hasAuthorship W3100877062A5084763603 @default.
- W3100877062 hasBestOaLocation W31008770622 @default.
- W3100877062 hasConcept C108583219 @default.
- W3100877062 hasConcept C111919701 @default.
- W3100877062 hasConcept C115961682 @default.
- W3100877062 hasConcept C152124472 @default.
- W3100877062 hasConcept C153180895 @default.
- W3100877062 hasConcept C154945302 @default.
- W3100877062 hasConcept C2776429412 @default.
- W3100877062 hasConcept C31972630 @default.
- W3100877062 hasConcept C36464697 @default.
- W3100877062 hasConcept C41008148 @default.
- W3100877062 hasConcept C52622490 @default.
- W3100877062 hasConcept C81363708 @default.
- W3100877062 hasConceptScore W3100877062C108583219 @default.
- W3100877062 hasConceptScore W3100877062C111919701 @default.
- W3100877062 hasConceptScore W3100877062C115961682 @default.