Matches in SemOpenAlex for { <https://semopenalex.org/work/W4319993407> ?p ?o ?g. }
Showing items 1 to 65 of
65
with 100 items per page.
- W4319993407 endingPage "14" @default.
- W4319993407 startingPage "1" @default.
- W4319993407 abstract "For pursuing accurate skeleton-based action recognition, most prior methods use the strategy of combining Graph Convolution Networks (GCNs) with attention-based methods in a serial way. However, they regard the human skeleton as a complete graph, resulting in less variations between different actions (e.g., the connection between the elbow and head in action “clapping hands”). For this, we propose a novel Contrastive GCN-Transformer Network (ConGT) which fuses the spatial and temporal modules in a parallel way. The ConGT involves two parallel streams: Spatial-Temporal Graph Convolution stream (STG) and Spatial-Temporal Transformer stream (STT). The STG is designed to obtain action representations maintaining the natural topology structure of the human skeleton. The STT is devised to acquire action representations containing the global relationships among joints. Since the action representations produced from these two streams contain different characteristics, and each of them knows little information of the other, we introduce the contrastive learning paradigm to guide their output representations of the same sample to be as close as possible in a self-supervised manner. Through the contrastive learning, they can learn information from each other to enrich the action features by maximizing the mutual information between the two types of action representations. To further improve action recognition accuracy, we introduce the Cyclical Focal Loss (CFL) which can focus on confident training samples in early training epochs, with an increasing focus on hard samples during the middle epochs. We conduct experiments on three benchmark datasets, which demonstrate that our model achieves state-of-the-art performance in action recognition." @default.
- W4319993407 created "2023-02-11" @default.
- W4319993407 creator A5001506992 @default.
- W4319993407 creator A5019288783 @default.
- W4319993407 creator A5077021587 @default.
- W4319993407 date "2023-01-01" @default.
- W4319993407 modified "2023-10-18" @default.
- W4319993407 title "Skeleton-Based Action Recognition Through Contrasting Two-Stream Spatial-Temporal Networks" @default.
- W4319993407 doi "https://doi.org/10.1109/tmm.2023.3239751" @default.
- W4319993407 hasPublicationYear "2023" @default.
- W4319993407 type Work @default.
- W4319993407 citedByCount "1" @default.
- W4319993407 countsByYear W43199934072023 @default.
- W4319993407 crossrefType "journal-article" @default.
- W4319993407 hasAuthorship W4319993407A5001506992 @default.
- W4319993407 hasAuthorship W4319993407A5019288783 @default.
- W4319993407 hasAuthorship W4319993407A5077021587 @default.
- W4319993407 hasConcept C120665830 @default.
- W4319993407 hasConcept C121332964 @default.
- W4319993407 hasConcept C132525143 @default.
- W4319993407 hasConcept C153180895 @default.
- W4319993407 hasConcept C154945302 @default.
- W4319993407 hasConcept C165801399 @default.
- W4319993407 hasConcept C192209626 @default.
- W4319993407 hasConcept C2777212361 @default.
- W4319993407 hasConcept C2987834672 @default.
- W4319993407 hasConcept C41008148 @default.
- W4319993407 hasConcept C45347329 @default.
- W4319993407 hasConcept C50644808 @default.
- W4319993407 hasConcept C62520636 @default.
- W4319993407 hasConcept C66322947 @default.
- W4319993407 hasConcept C80444323 @default.
- W4319993407 hasConceptScore W4319993407C120665830 @default.
- W4319993407 hasConceptScore W4319993407C121332964 @default.
- W4319993407 hasConceptScore W4319993407C132525143 @default.
- W4319993407 hasConceptScore W4319993407C153180895 @default.
- W4319993407 hasConceptScore W4319993407C154945302 @default.
- W4319993407 hasConceptScore W4319993407C165801399 @default.
- W4319993407 hasConceptScore W4319993407C192209626 @default.
- W4319993407 hasConceptScore W4319993407C2777212361 @default.
- W4319993407 hasConceptScore W4319993407C2987834672 @default.
- W4319993407 hasConceptScore W4319993407C41008148 @default.
- W4319993407 hasConceptScore W4319993407C45347329 @default.
- W4319993407 hasConceptScore W4319993407C50644808 @default.
- W4319993407 hasConceptScore W4319993407C62520636 @default.
- W4319993407 hasConceptScore W4319993407C66322947 @default.
- W4319993407 hasConceptScore W4319993407C80444323 @default.
- W4319993407 hasLocation W43199934071 @default.
- W4319993407 hasOpenAccess W4319993407 @default.
- W4319993407 hasPrimaryLocation W43199934071 @default.
- W4319993407 hasRelatedWork W1728708896 @default.
- W4319993407 hasRelatedWork W2726222394 @default.
- W4319993407 hasRelatedWork W2752217129 @default.
- W4319993407 hasRelatedWork W2929240682 @default.
- W4319993407 hasRelatedWork W2941155331 @default.
- W4319993407 hasRelatedWork W3106494386 @default.
- W4319993407 hasRelatedWork W4200171239 @default.
- W4319993407 hasRelatedWork W4226487993 @default.
- W4319993407 hasRelatedWork W4293731510 @default.
- W4319993407 hasRelatedWork W4312894178 @default.
- W4319993407 isParatext "false" @default.
- W4319993407 isRetracted "false" @default.
- W4319993407 workType "article" @default.