FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2962869397> ?p ?o ?g. }

• W2962869397 endingPage "98386" @default.
• W2962869397 startingPage "98374" @default.
• W2962869397 abstract "Currently, there are many works exploring how to more fully and efficiently use multi-scale feature maps of deep convolutional neural networks to improve the performance of object detection. But most of these works are devoted to predicting, respectively, on multi-scale feature maps or blending multi-scale feature maps for enriching representation. In this paper, we present a new method of cross fusing feature, named multi-semantic pyramids (MSP), for detecting different-scale objects. Various scale objects are predicted, respectively, by corresponding semantic pyramids (SP), each SP can produce rich semantic features for predicting via reusing inherent multi-scale feature maps from the network backbone. Through promoting the reuse of inherent feature layers, our MSP can improve detection performance with marginal extra cost. In addition, since the reuse connection of the MSP facilitates the conduction of the gradient, the convergence of the network is greatly improved. The experimental results on the PASCAL VOC and COCO datasets illustrate that our MSP can achieve more competitive detection accuracy." @default.
• W2962869397 created "2019-07-30" @default.
• W2962869397 creator A5012438499 @default.
• W2962869397 creator A5055643899 @default.
• W2962869397 creator A5087705980 @default.
• W2962869397 creator A5087786852 @default.
• W2962869397 date "2019-01-01" @default.
• W2962869397 modified "2023-10-12" @default.
• W2962869397 title "Fast Multi Semantic Pyramids via Cross Fusing Inherent Features for Different-Scale Detection" @default.
• W2962869397 cites W104184427 @default.
• W2962869397 cites W124653583 @default.
• W2962869397 cites W1536680647 @default.
• W2962869397 cites W1610060839 @default.
• W2962869397 cites W1686810756 @default.
• W2962869397 cites W1817277359 @default.
• W2962869397 cites W1836465849 @default.
• W2962869397 cites W1849277567 @default.
• W2962869397 cites W1861492603 @default.
• W2962869397 cites W1903029394 @default.
• W2962869397 cites W1948751323 @default.
• W2962869397 cites W2031489346 @default.
• W2962869397 cites W2056025798 @default.
• W2962869397 cites W2068730032 @default.
• W2962869397 cites W2097117768 @default.
• W2962869397 cites W2102605133 @default.
• W2962869397 cites W2112796928 @default.
• W2962869397 cites W2117539524 @default.
• W2962869397 cites W2122146326 @default.
• W2962869397 cites W2147800946 @default.
• W2962869397 cites W2155541015 @default.
• W2962869397 cites W2156387975 @default.
• W2962869397 cites W2163605009 @default.
• W2962869397 cites W2179352600 @default.
• W2962869397 cites W2194775991 @default.
• W2962869397 cites W2288122362 @default.
• W2962869397 cites W2304648132 @default.
• W2962869397 cites W2331143823 @default.
• W2962869397 cites W2490270993 @default.
• W2962869397 cites W2556967412 @default.
• W2962869397 cites W2565639579 @default.
• W2962869397 cites W2570343428 @default.
• W2962869397 cites W2579985080 @default.
• W2962869397 cites W2612445135 @default.
• W2962869397 cites W2613718673 @default.
• W2962869397 cites W2738890255 @default.
• W2962869397 cites W2796347433 @default.
• W2962869397 cites W2808910047 @default.
• W2962869397 cites W2810193307 @default.
• W2962869397 cites W2884561390 @default.
• W2962869397 cites W2902485530 @default.
• W2962869397 cites W2902609671 @default.
• W2962869397 cites W2903153723 @default.
• W2962869397 cites W2903268680 @default.
• W2962869397 cites W2903436395 @default.
• W2962869397 cites W2962721361 @default.
• W2962869397 cites W2962992847 @default.
• W2962869397 cites W2963032190 @default.
• W2962869397 cites W2963037989 @default.
• W2962869397 cites W2963113370 @default.
• W2962869397 cites W2963125010 @default.
• W2962869397 cites W2963287324 @default.
• W2962869397 cites W2963299740 @default.
• W2962869397 cites W2963307811 @default.
• W2962869397 cites W2963446712 @default.
• W2962869397 cites W2963813458 @default.
• W2962869397 cites W2963975324 @default.
• W2962869397 cites W2964288706 @default.
• W2962869397 cites W3106250896 @default.
• W2962869397 doi "https://doi.org/10.1109/access.2019.2930083" @default.
• W2962869397 hasPublicationYear "2019" @default.
• W2962869397 type Work @default.
• W2962869397 sameAs 2962869397 @default.
• W2962869397 citedByCount "1" @default.
• W2962869397 countsByYear W29628693972021 @default.
• W2962869397 crossrefType "journal-article" @default.
• W2962869397 hasAuthorship W2962869397A5012438499 @default.
• W2962869397 hasAuthorship W2962869397A5055643899 @default.
• W2962869397 hasAuthorship W2962869397A5087705980 @default.
• W2962869397 hasAuthorship W2962869397A5087786852 @default.
• W2962869397 hasBestOaLocation W29628693971 @default.
• W2962869397 hasConcept C121332964 @default.
• W2962869397 hasConcept C124101348 @default.
• W2962869397 hasConcept C127413603 @default.
• W2962869397 hasConcept C138885662 @default.
• W2962869397 hasConcept C153180895 @default.
• W2962869397 hasConcept C154945302 @default.
• W2962869397 hasConcept C199360897 @default.
• W2962869397 hasConcept C206588197 @default.
• W2962869397 hasConcept C2776151529 @default.
• W2962869397 hasConcept C2776401178 @default.
• W2962869397 hasConcept C2778755073 @default.
• W2962869397 hasConcept C2781122975 @default.
• W2962869397 hasConcept C31258907 @default.
• W2962869397 hasConcept C41008148 @default.
• W2962869397 hasConcept C41895202 @default.
• W2962869397 hasConcept C548081761 @default.
• W2962869397 hasConcept C62520636 @default.
• W2962869397 hasConcept C75608658 @default.

• next