Matches in SemOpenAlex for { <https://semopenalex.org/work/W2900073955> ?p ?o ?g. }
Showing items 1 to 90 of
90
with 100 items per page.
- W2900073955 endingPage "68633" @default.
- W2900073955 startingPage "68621" @default.
- W2900073955 abstract "Improving the performance of CNN-based mobile applications by offloading its computation from mobile devices to the cloud has attracted the attention of the community. Generally, there are three stages in the workflow, including local inference on the mobile device, data transmission of the intermediate result, and remote inference in the cloud. However, the time cost of local inference and data transmission are still the bottleneck in reaching the desirable inference performance. In this paper, we propose an image-aware inference framework called IF-CNN to enable fast inference based on computation offloading. In the framework, we first build a model pool consisting of CNN models with different complexities. The most efficient one from such candidate models is selected to process the corresponding image. During the selection process, we have designed an effective model to predict the confidence based on multi-task learning. After model selection, half-floating optimization and feature compression are applied to accelerate the process of distributed inference between mobile devices and cloud. Experimental results show that IF-CNN is credible to identify the most effective model for different images and the total inference performance could be significantly improved. Meanwhile, IF-CNN is complementary to other inference acceleration methods of CNN models." @default.
- W2900073955 created "2018-11-16" @default.
- W2900073955 creator A5018342513 @default.
- W2900073955 creator A5043376108 @default.
- W2900073955 creator A5078020203 @default.
- W2900073955 creator A5078642887 @default.
- W2900073955 date "2018-01-01" @default.
- W2900073955 modified "2023-09-29" @default.
- W2900073955 title "IF-CNN: Image-Aware Inference Framework for CNN With the Collaboration of Mobile Devices and Cloud" @default.
- W2900073955 cites W1669690275 @default.
- W2900073955 cites W2012215713 @default.
- W2900073955 cites W2012423440 @default.
- W2900073955 cites W2023380813 @default.
- W2900073955 cites W2078224158 @default.
- W2900073955 cites W2086754606 @default.
- W2900073955 cites W2088692353 @default.
- W2900073955 cites W2101788345 @default.
- W2900073955 cites W2113120414 @default.
- W2900073955 cites W2115514375 @default.
- W2900073955 cites W2118016489 @default.
- W2900073955 cites W2143419558 @default.
- W2900073955 cites W2148254922 @default.
- W2900073955 cites W2155893237 @default.
- W2900073955 cites W2233116163 @default.
- W2900073955 cites W2337472605 @default.
- W2900073955 cites W2529546376 @default.
- W2900073955 cites W2531409750 @default.
- W2900073955 cites W2605258629 @default.
- W2900073955 cites W2625759119 @default.
- W2900073955 cites W2626129225 @default.
- W2900073955 cites W2742405391 @default.
- W2900073955 cites W2760063890 @default.
- W2900073955 cites W2963037989 @default.
- W2900073955 cites W2963770864 @default.
- W2900073955 cites W2963828468 @default.
- W2900073955 doi "https://doi.org/10.1109/access.2018.2880196" @default.
- W2900073955 hasPublicationYear "2018" @default.
- W2900073955 type Work @default.
- W2900073955 sameAs 2900073955 @default.
- W2900073955 citedByCount "20" @default.
- W2900073955 countsByYear W29000739552019 @default.
- W2900073955 countsByYear W29000739552020 @default.
- W2900073955 countsByYear W29000739552021 @default.
- W2900073955 countsByYear W29000739552022 @default.
- W2900073955 countsByYear W29000739552023 @default.
- W2900073955 crossrefType "journal-article" @default.
- W2900073955 hasAuthorship W2900073955A5018342513 @default.
- W2900073955 hasAuthorship W2900073955A5043376108 @default.
- W2900073955 hasAuthorship W2900073955A5078020203 @default.
- W2900073955 hasAuthorship W2900073955A5078642887 @default.
- W2900073955 hasBestOaLocation W29000739551 @default.
- W2900073955 hasConcept C111919701 @default.
- W2900073955 hasConcept C115961682 @default.
- W2900073955 hasConcept C136764020 @default.
- W2900073955 hasConcept C154945302 @default.
- W2900073955 hasConcept C186967261 @default.
- W2900073955 hasConcept C2776214188 @default.
- W2900073955 hasConcept C41008148 @default.
- W2900073955 hasConcept C79974875 @default.
- W2900073955 hasConcept C81363708 @default.
- W2900073955 hasConceptScore W2900073955C111919701 @default.
- W2900073955 hasConceptScore W2900073955C115961682 @default.
- W2900073955 hasConceptScore W2900073955C136764020 @default.
- W2900073955 hasConceptScore W2900073955C154945302 @default.
- W2900073955 hasConceptScore W2900073955C186967261 @default.
- W2900073955 hasConceptScore W2900073955C2776214188 @default.
- W2900073955 hasConceptScore W2900073955C41008148 @default.
- W2900073955 hasConceptScore W2900073955C79974875 @default.
- W2900073955 hasConceptScore W2900073955C81363708 @default.
- W2900073955 hasFunder F4320321540 @default.
- W2900073955 hasLocation W29000739551 @default.
- W2900073955 hasOpenAccess W2900073955 @default.
- W2900073955 hasPrimaryLocation W29000739551 @default.
- W2900073955 hasRelatedWork W2383532021 @default.
- W2900073955 hasRelatedWork W2810384904 @default.
- W2900073955 hasRelatedWork W3114074529 @default.
- W2900073955 hasRelatedWork W4200370950 @default.
- W2900073955 hasRelatedWork W4221146321 @default.
- W2900073955 hasRelatedWork W4239686595 @default.
- W2900073955 hasRelatedWork W4289229149 @default.
- W2900073955 hasRelatedWork W4321349613 @default.
- W2900073955 hasRelatedWork W4366818321 @default.
- W2900073955 hasRelatedWork W3090693157 @default.
- W2900073955 hasVolume "6" @default.
- W2900073955 isParatext "false" @default.
- W2900073955 isRetracted "false" @default.
- W2900073955 magId "2900073955" @default.
- W2900073955 workType "article" @default.