Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313150900> ?p ?o ?g. }
- W4313150900 endingPage "3008" @default.
- W4313150900 startingPage "2994" @default.
- W4313150900 abstract "Generative adversarial networks (GANs) have recently been proposed for the synthesis of RF micro-Doppler signatures to address the issue of low sample support and enable the training of deeper neural networks (DNNs) for enhanced RF signal classification. But GANs suffer from systemic kinematic inconsistencies that decrease performance when GAN-synthesized data is used for training DNNs in human activity recognition. As a solution to this problem, this article proposes the design of a multibranch GAN (MBGAN), which integrates domain knowledge into its architecture, and physics-aware metrics based on correlation and curve-matching in the loss function. The quality of the synthetic samples generated is evaluated via image quality metrics, the ability to synthesize data that reflects human physical properties and generalize to broader subject profiles, and the achieved classification accuracy. Our experimental results show the proposed approach generates synthetic data for training that more accurately matches target kinematics, resulting in an increase of 9% in classification accuracy when classifying 14 different ambulatory human activities." @default.
- W4313150900 created "2023-01-06" @default.
- W4313150900 creator A5007661434 @default.
- W4313150900 creator A5025392795 @default.
- W4313150900 creator A5076081220 @default.
- W4313150900 date "2023-06-01" @default.
- W4313150900 modified "2023-10-18" @default.
- W4313150900 title "Physics-Aware Generative Adversarial Networks for Radar-Based Human Activity Recognition" @default.
- W4313150900 cites W132064424 @default.
- W4313150900 cites W1500831240 @default.
- W4313150900 cites W2020473868 @default.
- W4313150900 cites W2075341176 @default.
- W4313150900 cites W2125101937 @default.
- W4313150900 cites W2132481658 @default.
- W4313150900 cites W2133665775 @default.
- W4313150900 cites W2148832324 @default.
- W4313150900 cites W2154884741 @default.
- W4313150900 cites W2207110500 @default.
- W4313150900 cites W2249376573 @default.
- W4313150900 cites W2501169257 @default.
- W4313150900 cites W2626065224 @default.
- W4313150900 cites W2891278391 @default.
- W4313150900 cites W2899283552 @default.
- W4313150900 cites W2900369848 @default.
- W4313150900 cites W2931068004 @default.
- W4313150900 cites W2944605902 @default.
- W4313150900 cites W2946794331 @default.
- W4313150900 cites W2953001150 @default.
- W4313150900 cites W2954377489 @default.
- W4313150900 cites W2973630397 @default.
- W4313150900 cites W2983431682 @default.
- W4313150900 cites W2993835312 @default.
- W4313150900 cites W3003598450 @default.
- W4313150900 cites W3006839063 @default.
- W4313150900 cites W3018508273 @default.
- W4313150900 cites W3034433182 @default.
- W4313150900 cites W3035427195 @default.
- W4313150900 cites W3083870581 @default.
- W4313150900 cites W3096831136 @default.
- W4313150900 cites W3098389423 @default.
- W4313150900 cites W3104772982 @default.
- W4313150900 cites W3136270431 @default.
- W4313150900 cites W3163242382 @default.
- W4313150900 cites W3163993681 @default.
- W4313150900 cites W3173403881 @default.
- W4313150900 cites W3179793697 @default.
- W4313150900 cites W3181354245 @default.
- W4313150900 cites W3207471124 @default.
- W4313150900 cites W4200581268 @default.
- W4313150900 cites W4206328226 @default.
- W4313150900 cites W4226073445 @default.
- W4313150900 cites W4230707046 @default.
- W4313150900 cites W4293860566 @default.
- W4313150900 cites W2083106287 @default.
- W4313150900 doi "https://doi.org/10.1109/taes.2022.3221023" @default.
- W4313150900 hasPublicationYear "2023" @default.
- W4313150900 type Work @default.
- W4313150900 citedByCount "2" @default.
- W4313150900 countsByYear W43131509002022 @default.
- W4313150900 countsByYear W43131509002023 @default.
- W4313150900 crossrefType "journal-article" @default.
- W4313150900 hasAuthorship W4313150900A5007661434 @default.
- W4313150900 hasAuthorship W4313150900A5025392795 @default.
- W4313150900 hasAuthorship W4313150900A5076081220 @default.
- W4313150900 hasConcept C105795698 @default.
- W4313150900 hasConcept C119857082 @default.
- W4313150900 hasConcept C121332964 @default.
- W4313150900 hasConcept C14036430 @default.
- W4313150900 hasConcept C153180895 @default.
- W4313150900 hasConcept C154945302 @default.
- W4313150900 hasConcept C165064840 @default.
- W4313150900 hasConcept C199360897 @default.
- W4313150900 hasConcept C2779843651 @default.
- W4313150900 hasConcept C33923547 @default.
- W4313150900 hasConcept C37736160 @default.
- W4313150900 hasConcept C39890363 @default.
- W4313150900 hasConcept C39920418 @default.
- W4313150900 hasConcept C41008148 @default.
- W4313150900 hasConcept C50644808 @default.
- W4313150900 hasConcept C554190296 @default.
- W4313150900 hasConcept C74650414 @default.
- W4313150900 hasConcept C76155785 @default.
- W4313150900 hasConcept C78458016 @default.
- W4313150900 hasConcept C86803240 @default.
- W4313150900 hasConceptScore W4313150900C105795698 @default.
- W4313150900 hasConceptScore W4313150900C119857082 @default.
- W4313150900 hasConceptScore W4313150900C121332964 @default.
- W4313150900 hasConceptScore W4313150900C14036430 @default.
- W4313150900 hasConceptScore W4313150900C153180895 @default.
- W4313150900 hasConceptScore W4313150900C154945302 @default.
- W4313150900 hasConceptScore W4313150900C165064840 @default.
- W4313150900 hasConceptScore W4313150900C199360897 @default.
- W4313150900 hasConceptScore W4313150900C2779843651 @default.
- W4313150900 hasConceptScore W4313150900C33923547 @default.
- W4313150900 hasConceptScore W4313150900C37736160 @default.
- W4313150900 hasConceptScore W4313150900C39890363 @default.
- W4313150900 hasConceptScore W4313150900C39920418 @default.
- W4313150900 hasConceptScore W4313150900C41008148 @default.