Matches in SemOpenAlex for { <https://semopenalex.org/work/W4361762277> ?p ?o ?g. }
- W4361762277 endingPage "5946" @default.
- W4361762277 startingPage "5932" @default.
- W4361762277 abstract "Few-shot action recognition has attracted increasing attention in recent years, but it remains challenging due to the intrinsic difficulty in learning transferable knowledge to generalize to novel classes by using a few labeled samples. Although some successful progress has been made, most few-shot action recognition methods commonly focus on the global characteristics of samples while ignoring the local characteristics of samples, which results in the weak generalization ability of the model. In this paper, we propose a task-aware dual-representation network (TADRNet) for few-shot action recognition, which learns how to adapt video representations to novel tasks in a meta-learning manner. It mainly includes a global relational graph subnetwork (GRG) and a fine-grained local representation subnetwork (FLR). Our method simultaneously considers both global and local characteristics of samples for few-shot action recognition. From a global perspective, we propose GRG to explore the relations across support-query sample pairs by using the relational graph neural network. To facilitate the few-shot visual learning, we propose a novel hybrid semantic attention module (HSA) for enhancing the discriminability of support and query features. From a local perspective, we utilize FLR to fully exploit the local characteristics of samples, which can improve the classification results obtained by GRG and thus guarantee high classification accuracy. Extensive experiments on four challenging benchmarks show that the proposed TADRNet significantly outperforms a variety of state-of-the-art few-shot action recognition methods." @default.
- W4361762277 created "2023-04-04" @default.
- W4361762277 creator A5001949272 @default.
- W4361762277 creator A5015095434 @default.
- W4361762277 creator A5032198937 @default.
- W4361762277 creator A5044594971 @default.
- W4361762277 creator A5044972184 @default.
- W4361762277 creator A5058010200 @default.
- W4361762277 creator A5072608526 @default.
- W4361762277 creator A5088475144 @default.
- W4361762277 date "2023-10-01" @default.
- W4361762277 modified "2023-10-05" @default.
- W4361762277 title "Task-Aware Dual-Representation Network for Few-Shot Action Recognition" @default.
- W4361762277 cites W1522734439 @default.
- W4361762277 cites W2108598243 @default.
- W4361762277 cites W2115733720 @default.
- W4361762277 cites W2126579184 @default.
- W4361762277 cites W2166742463 @default.
- W4361762277 cites W2194775991 @default.
- W4361762277 cites W2342662179 @default.
- W4361762277 cites W2507009361 @default.
- W4361762277 cites W2625366777 @default.
- W4361762277 cites W2751206209 @default.
- W4361762277 cites W2770804203 @default.
- W4361762277 cites W2791303549 @default.
- W4361762277 cites W2883429621 @default.
- W4361762277 cites W2894873912 @default.
- W4361762277 cites W2942750779 @default.
- W4361762277 cites W2963070905 @default.
- W4361762277 cites W2963155035 @default.
- W4361762277 cites W2963524571 @default.
- W4361762277 cites W2964105864 @default.
- W4361762277 cites W2979689312 @default.
- W4361762277 cites W2981874246 @default.
- W4361762277 cites W2990152177 @default.
- W4361762277 cites W2990503944 @default.
- W4361762277 cites W2994785205 @default.
- W4361762277 cites W3009514405 @default.
- W4361762277 cites W3034187513 @default.
- W4361762277 cites W3034768625 @default.
- W4361762277 cites W3035163205 @default.
- W4361762277 cites W3035254087 @default.
- W4361762277 cites W3035303837 @default.
- W4361762277 cites W3035374961 @default.
- W4361762277 cites W3047772167 @default.
- W4361762277 cites W3093455342 @default.
- W4361762277 cites W3095374178 @default.
- W4361762277 cites W3168565360 @default.
- W4361762277 cites W3173271747 @default.
- W4361762277 cites W3173685130 @default.
- W4361762277 cites W3175357208 @default.
- W4361762277 cites W3175528717 @default.
- W4361762277 cites W3183316215 @default.
- W4361762277 cites W3187343781 @default.
- W4361762277 cites W3190693140 @default.
- W4361762277 cites W3190960741 @default.
- W4361762277 cites W3205333272 @default.
- W4361762277 cites W3205808193 @default.
- W4361762277 cites W3206402863 @default.
- W4361762277 cites W3206930557 @default.
- W4361762277 cites W3207395111 @default.
- W4361762277 cites W3208210572 @default.
- W4361762277 cites W4285184936 @default.
- W4361762277 cites W4312959318 @default.
- W4361762277 cites W4313046672 @default.
- W4361762277 doi "https://doi.org/10.1109/tcsvt.2023.3262670" @default.
- W4361762277 hasPublicationYear "2023" @default.
- W4361762277 type Work @default.
- W4361762277 citedByCount "0" @default.
- W4361762277 crossrefType "journal-article" @default.
- W4361762277 hasAuthorship W4361762277A5001949272 @default.
- W4361762277 hasAuthorship W4361762277A5015095434 @default.
- W4361762277 hasAuthorship W4361762277A5032198937 @default.
- W4361762277 hasAuthorship W4361762277A5044594971 @default.
- W4361762277 hasAuthorship W4361762277A5044972184 @default.
- W4361762277 hasAuthorship W4361762277A5058010200 @default.
- W4361762277 hasAuthorship W4361762277A5072608526 @default.
- W4361762277 hasAuthorship W4361762277A5088475144 @default.
- W4361762277 hasConcept C119857082 @default.
- W4361762277 hasConcept C124952713 @default.
- W4361762277 hasConcept C12713177 @default.
- W4361762277 hasConcept C132525143 @default.
- W4361762277 hasConcept C134306372 @default.
- W4361762277 hasConcept C142362112 @default.
- W4361762277 hasConcept C153180895 @default.
- W4361762277 hasConcept C154945302 @default.
- W4361762277 hasConcept C162324750 @default.
- W4361762277 hasConcept C177148314 @default.
- W4361762277 hasConcept C17744445 @default.
- W4361762277 hasConcept C187736073 @default.
- W4361762277 hasConcept C199539241 @default.
- W4361762277 hasConcept C2776359362 @default.
- W4361762277 hasConcept C2780186347 @default.
- W4361762277 hasConcept C2780451532 @default.
- W4361762277 hasConcept C2780980858 @default.
- W4361762277 hasConcept C33923547 @default.
- W4361762277 hasConcept C38652104 @default.
- W4361762277 hasConcept C41008148 @default.