Matches in SemOpenAlex for { <https://semopenalex.org/work/W2936599103> ?p ?o ?g. }
- W2936599103 abstract "Current state-of-the-art convolutional architectures for object detection are manually designed. Here we aim to learn a better architecture of feature pyramid network for object detection. We adopt Neural Architecture Search and discover a new feature pyramid architecture in a novel scalable search space covering all cross-scale connections. The discovered architecture, named NAS-FPN, consists of a combination of top-down and bottom-up connections to fuse features across scales. NAS-FPN, combined with various backbone models in the RetinaNet framework, achieves better accuracy and latency tradeoff compared to state-of-the-art object detection models. NAS-FPN improves mobile detection accuracy by 2 AP compared to state-of-the-art SSDLite with MobileNetV2 model in [32] and achieves 48.3 AP which surpasses Mask R-CNN [10] detection accuracy with less computation time." @default.
- W2936599103 created "2019-04-25" @default.
- W2936599103 creator A5002729731 @default.
- W2936599103 creator A5029917821 @default.
- W2936599103 creator A5052768778 @default.
- W2936599103 creator A5088551093 @default.
- W2936599103 date "2019-04-16" @default.
- W2936599103 modified "2023-09-23" @default.
- W2936599103 title "NAS-FPN: Learning Scalable Feature Pyramid Architecture for Object Detection" @default.
- W2936599103 cites W1901129140 @default.
- W2936599103 cites W2508741746 @default.
- W2936599103 cites W2511730936 @default.
- W2936599103 cites W2553303224 @default.
- W2936599103 cites W2579985080 @default.
- W2936599103 cites W2610140147 @default.
- W2936599103 cites W2736601468 @default.
- W2936599103 cites W2736941579 @default.
- W2936599103 cites W2738535327 @default.
- W2936599103 cites W2743473392 @default.
- W2936599103 cites W2755164689 @default.
- W2936599103 cites W2771727678 @default.
- W2936599103 cites W2793693263 @default.
- W2936599103 cites W2796347433 @default.
- W2936599103 cites W2797527871 @default.
- W2936599103 cites W2885311373 @default.
- W2936599103 cites W2886335102 @default.
- W2936599103 cites W2890166761 @default.
- W2936599103 cites W2891778567 @default.
- W2936599103 cites W2949533892 @default.
- W2936599103 cites W2949650786 @default.
- W2936599103 cites W2949658958 @default.
- W2936599103 cites W2950762923 @default.
- W2936599103 cites W2951886768 @default.
- W2936599103 cites W2962917547 @default.
- W2936599103 cites W2963163009 @default.
- W2936599103 cites W2964143949 @default.
- W2936599103 cites W2997225633 @default.
- W2936599103 cites W8437397 @default.
- W2936599103 hasPublicationYear "2019" @default.
- W2936599103 type Work @default.
- W2936599103 sameAs 2936599103 @default.
- W2936599103 citedByCount "153" @default.
- W2936599103 countsByYear W29365991032019 @default.
- W2936599103 countsByYear W29365991032020 @default.
- W2936599103 countsByYear W29365991032021 @default.
- W2936599103 countsByYear W29365991032022 @default.
- W2936599103 crossrefType "posted-content" @default.
- W2936599103 hasAuthorship W2936599103A5002729731 @default.
- W2936599103 hasAuthorship W2936599103A5029917821 @default.
- W2936599103 hasAuthorship W2936599103A5052768778 @default.
- W2936599103 hasAuthorship W2936599103A5088551093 @default.
- W2936599103 hasConcept C11413529 @default.
- W2936599103 hasConcept C119599485 @default.
- W2936599103 hasConcept C123657996 @default.
- W2936599103 hasConcept C127413603 @default.
- W2936599103 hasConcept C138885662 @default.
- W2936599103 hasConcept C141353440 @default.
- W2936599103 hasConcept C142575187 @default.
- W2936599103 hasConcept C153180895 @default.
- W2936599103 hasConcept C154945302 @default.
- W2936599103 hasConcept C166957645 @default.
- W2936599103 hasConcept C205649164 @default.
- W2936599103 hasConcept C2524010 @default.
- W2936599103 hasConcept C2776151529 @default.
- W2936599103 hasConcept C2776401178 @default.
- W2936599103 hasConcept C2781238097 @default.
- W2936599103 hasConcept C31972630 @default.
- W2936599103 hasConcept C33923547 @default.
- W2936599103 hasConcept C41008148 @default.
- W2936599103 hasConcept C41895202 @default.
- W2936599103 hasConcept C45374587 @default.
- W2936599103 hasConcept C48044578 @default.
- W2936599103 hasConcept C76155785 @default.
- W2936599103 hasConcept C77088390 @default.
- W2936599103 hasConcept C81363708 @default.
- W2936599103 hasConcept C82876162 @default.
- W2936599103 hasConceptScore W2936599103C11413529 @default.
- W2936599103 hasConceptScore W2936599103C119599485 @default.
- W2936599103 hasConceptScore W2936599103C123657996 @default.
- W2936599103 hasConceptScore W2936599103C127413603 @default.
- W2936599103 hasConceptScore W2936599103C138885662 @default.
- W2936599103 hasConceptScore W2936599103C141353440 @default.
- W2936599103 hasConceptScore W2936599103C142575187 @default.
- W2936599103 hasConceptScore W2936599103C153180895 @default.
- W2936599103 hasConceptScore W2936599103C154945302 @default.
- W2936599103 hasConceptScore W2936599103C166957645 @default.
- W2936599103 hasConceptScore W2936599103C205649164 @default.
- W2936599103 hasConceptScore W2936599103C2524010 @default.
- W2936599103 hasConceptScore W2936599103C2776151529 @default.
- W2936599103 hasConceptScore W2936599103C2776401178 @default.
- W2936599103 hasConceptScore W2936599103C2781238097 @default.
- W2936599103 hasConceptScore W2936599103C31972630 @default.
- W2936599103 hasConceptScore W2936599103C33923547 @default.
- W2936599103 hasConceptScore W2936599103C41008148 @default.
- W2936599103 hasConceptScore W2936599103C41895202 @default.
- W2936599103 hasConceptScore W2936599103C45374587 @default.
- W2936599103 hasConceptScore W2936599103C48044578 @default.
- W2936599103 hasConceptScore W2936599103C76155785 @default.
- W2936599103 hasConceptScore W2936599103C77088390 @default.
- W2936599103 hasConceptScore W2936599103C81363708 @default.