FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2893890202> ?p ?o ?g. }

• W2893890202 abstract "We consider an important task of effective and efficient semantic image segmentation. In particular, we adapt a powerful semantic segmentation architecture, called RefineNet, into the more compact one, suitable even for tasks requiring real-time performance on high-resolution inputs. To this end, we identify computationally expensive blocks in the original setup, and propose two modifications aimed to decrease the number of parameters and floating point operations. By doing that, we achieve more than twofold model reduction, while keeping the performance levels almost intact. Our fastest model undergoes a significant speed-up boost from 20 FPS to 55 FPS on a generic GPU card on 512x512 inputs with solid 81.1% mean iou performance on the test set of PASCAL VOC, while our slowest model with 32 FPS (from original 17 FPS) shows 82.7% mean iou on the same dataset. Alternatively, we showcase that our approach is easily mixable with light-weight classification networks: we attain 79.2% mean iou on PASCAL VOC using a model that contains only 3.3M parameters and performs only 9.3B floating point operations." @default.
• W2893890202 created "2018-10-05" @default.
• W2893890202 creator A5006294869 @default.
• W2893890202 creator A5087453555 @default.
• W2893890202 creator A5089444805 @default.
• W2893890202 date "2018-10-08" @default.
• W2893890202 modified "2023-10-06" @default.
• W2893890202 title "Light-Weight RefineNet for Real-Time Semantic Segmentation" @default.
• W2893890202 cites W125693051 @default.
• W2893890202 cites W1487583988 @default.
• W2893890202 cites W1514535095 @default.
• W2893890202 cites W1528789833 @default.
• W2893890202 cites W1686810756 @default.
• W2893890202 cites W1724438581 @default.
• W2893890202 cites W1821462560 @default.
• W2893890202 cites W1849277567 @default.
• W2893890202 cites W1901129140 @default.
• W2893890202 cites W1903029394 @default.
• W2893890202 cites W1905882502 @default.
• W2893890202 cites W1923697677 @default.
• W2893890202 cites W2031489346 @default.
• W2893890202 cites W2067912884 @default.
• W2893890202 cites W2100588357 @default.
• W2893890202 cites W2108598243 @default.
• W2893890202 cites W2114766824 @default.
• W2893890202 cites W2119144962 @default.
• W2893890202 cites W2125215748 @default.
• W2893890202 cites W2125389748 @default.
• W2893890202 cites W2134797427 @default.
• W2893890202 cites W2151103935 @default.
• W2893890202 cites W2158865742 @default.
• W2893890202 cites W2161969291 @default.
• W2893890202 cites W2167215970 @default.
• W2893890202 cites W2167510172 @default.
• W2893890202 cites W2194775991 @default.
• W2893890202 cites W2279098554 @default.
• W2893890202 cites W2300770211 @default.
• W2893890202 cites W2419448466 @default.
• W2893890202 cites W2531409750 @default.
• W2893890202 cites W2535516436 @default.
• W2893890202 cites W2545985378 @default.
• W2893890202 cites W2559767995 @default.
• W2893890202 cites W2560023338 @default.
• W2893890202 cites W2575705757 @default.
• W2893890202 cites W2593245696 @default.
• W2893890202 cites W2612445135 @default.
• W2893890202 cites W2619538244 @default.
• W2893890202 cites W2630837129 @default.
• W2893890202 cites W2734663976 @default.
• W2893890202 cites W2736728583 @default.
• W2893890202 cites W2736941579 @default.
• W2893890202 cites W2787091153 @default.
• W2893890202 cites W2899771611 @default.
• W2893890202 cites W2949117887 @default.
• W2893890202 cites W2949382160 @default.
• W2893890202 cites W2950967261 @default.
• W2893890202 cites W2951329458 @default.
• W2893890202 cites W2951638509 @default.
• W2893890202 cites W2952865063 @default.
• W2893890202 cites W2953106684 @default.
• W2893890202 cites W2953139137 @default.
• W2893890202 cites W2963446712 @default.
• W2893890202 cites W2963674932 @default.
• W2893890202 cites W360623563 @default.
• W2893890202 cites W73112891 @default.
• W2893890202 hasPublicationYear "2018" @default.
• W2893890202 type Work @default.
• W2893890202 sameAs 2893890202 @default.
• W2893890202 citedByCount "23" @default.
• W2893890202 countsByYear W28938902022018 @default.
• W2893890202 countsByYear W28938902022019 @default.
• W2893890202 countsByYear W28938902022020 @default.
• W2893890202 countsByYear W28938902022021 @default.
• W2893890202 crossrefType "posted-content" @default.
• W2893890202 hasAuthorship W2893890202A5006294869 @default.
• W2893890202 hasAuthorship W2893890202A5087453555 @default.
• W2893890202 hasAuthorship W2893890202A5089444805 @default.
• W2893890202 hasConcept C127413603 @default.
• W2893890202 hasConcept C153180895 @default.
• W2893890202 hasConcept C154945302 @default.
• W2893890202 hasConcept C199360897 @default.
• W2893890202 hasConcept C201995342 @default.
• W2893890202 hasConcept C2524010 @default.
• W2893890202 hasConcept C2780451532 @default.
• W2893890202 hasConcept C28719098 @default.
• W2893890202 hasConcept C33923547 @default.
• W2893890202 hasConcept C41008148 @default.
• W2893890202 hasConcept C75608658 @default.
• W2893890202 hasConcept C89600930 @default.
• W2893890202 hasConceptScore W2893890202C127413603 @default.
• W2893890202 hasConceptScore W2893890202C153180895 @default.
• W2893890202 hasConceptScore W2893890202C154945302 @default.
• W2893890202 hasConceptScore W2893890202C199360897 @default.
• W2893890202 hasConceptScore W2893890202C201995342 @default.
• W2893890202 hasConceptScore W2893890202C2524010 @default.
• W2893890202 hasConceptScore W2893890202C2780451532 @default.
• W2893890202 hasConceptScore W2893890202C28719098 @default.
• W2893890202 hasConceptScore W2893890202C33923547 @default.
• W2893890202 hasConceptScore W2893890202C41008148 @default.
• W2893890202 hasConceptScore W2893890202C75608658 @default.

• next