Matches in SemOpenAlex for { <https://semopenalex.org/work/W3209504420> ?p ?o ?g. }
- W3209504420 abstract "Recognizing human actions is fundamentally a spatio-temporal reasoning problem, and should be, at least to some extent, invariant to the appearance of the human and the objects involved. Motivated by this hypothesis, in this work, we take an object-centric approach to action recognition. Multiple works have studied this setting before, yet it remains unclear (i) how well a carefully crafted, spatio-temporal layout-based method can recognize human actions, and (ii) how, and when, to fuse the information from layout and appearance-based models. The main focus of this paper is compositional/few-shot action recognition, where we advocate the usage of multi-head attention (proven to be effective for spatial reasoning) over spatio-temporal layouts, i.e., configurations of object bounding boxes. We evaluate different schemes to inject video appearance information to the system, and benchmark our approach on background cluttered action recognition. On the Something-Else and Action Genome datasets, we demonstrate (i) how to extend multi-head attention for spatio-temporal layout-based action recognition, (ii) how to improve the performance of appearance-based models by fusion with layout-based models, (iii) that even on non-compositional background-cluttered video datasets, a fusion between layout- and appearance-based models improves the performance." @default.
- W3209504420 created "2021-11-08" @default.
- W3209504420 creator A5033652871 @default.
- W3209504420 creator A5074816094 @default.
- W3209504420 creator A5075796989 @default.
- W3209504420 date "2021-11-03" @default.
- W3209504420 modified "2023-09-25" @default.
- W3209504420 title "Revisiting spatio-temporal layouts for compositional action recognition." @default.
- W3209504420 cites W1539811621 @default.
- W3209504420 cites W1861492603 @default.
- W3209504420 cites W1933349210 @default.
- W3209504420 cites W2064675550 @default.
- W3209504420 cites W2095705004 @default.
- W3209504420 cites W2108598243 @default.
- W3209504420 cites W2112796928 @default.
- W3209504420 cites W2156303437 @default.
- W3209504420 cites W2157331557 @default.
- W3209504420 cites W2194775991 @default.
- W3209504420 cites W2337252826 @default.
- W3209504420 cites W2507009361 @default.
- W3209504420 cites W2613718673 @default.
- W3209504420 cites W2625366777 @default.
- W3209504420 cites W2770804203 @default.
- W3209504420 cites W2796742604 @default.
- W3209504420 cites W2806331055 @default.
- W3209504420 cites W2808675313 @default.
- W3209504420 cites W2883275382 @default.
- W3209504420 cites W2883429621 @default.
- W3209504420 cites W2888471892 @default.
- W3209504420 cites W2951529591 @default.
- W3209504420 cites W2955874753 @default.
- W3209504420 cites W2961193895 @default.
- W3209504420 cites W2962688385 @default.
- W3209504420 cites W2962711930 @default.
- W3209504420 cites W2962850006 @default.
- W3209504420 cites W2963091558 @default.
- W3209504420 cites W2963150697 @default.
- W3209504420 cites W2963155035 @default.
- W3209504420 cites W2963192057 @default.
- W3209504420 cites W2963403868 @default.
- W3209504420 cites W2963524571 @default.
- W3209504420 cites W2963563276 @default.
- W3209504420 cites W2963699792 @default.
- W3209504420 cites W2964015378 @default.
- W3209504420 cites W2970231061 @default.
- W3209504420 cites W2970608575 @default.
- W3209504420 cites W2981851019 @default.
- W3209504420 cites W2990152177 @default.
- W3209504420 cites W2990503944 @default.
- W3209504420 cites W2992457155 @default.
- W3209504420 cites W2994759459 @default.
- W3209504420 cites W3014411586 @default.
- W3209504420 cites W3015092156 @default.
- W3209504420 cites W3034257141 @default.
- W3209504420 cites W3034679267 @default.
- W3209504420 cites W3035413240 @default.
- W3209504420 cites W3096609285 @default.
- W3209504420 cites W3097065222 @default.
- W3209504420 cites W3100899490 @default.
- W3209504420 cites W3111098292 @default.
- W3209504420 cites W3119786062 @default.
- W3209504420 cites W3119997354 @default.
- W3209504420 cites W3126721948 @default.
- W3209504420 cites W3171660447 @default.
- W3209504420 hasPublicationYear "2021" @default.
- W3209504420 type Work @default.
- W3209504420 sameAs 3209504420 @default.
- W3209504420 citedByCount "0" @default.
- W3209504420 crossrefType "posted-content" @default.
- W3209504420 hasAuthorship W3209504420A5033652871 @default.
- W3209504420 hasAuthorship W3209504420A5074816094 @default.
- W3209504420 hasAuthorship W3209504420A5075796989 @default.
- W3209504420 hasConcept C119599485 @default.
- W3209504420 hasConcept C119857082 @default.
- W3209504420 hasConcept C120665830 @default.
- W3209504420 hasConcept C121332964 @default.
- W3209504420 hasConcept C127413603 @default.
- W3209504420 hasConcept C13280743 @default.
- W3209504420 hasConcept C141353440 @default.
- W3209504420 hasConcept C153180895 @default.
- W3209504420 hasConcept C154945302 @default.
- W3209504420 hasConcept C185798385 @default.
- W3209504420 hasConcept C190470478 @default.
- W3209504420 hasConcept C192209626 @default.
- W3209504420 hasConcept C205649164 @default.
- W3209504420 hasConcept C2777212361 @default.
- W3209504420 hasConcept C2780791683 @default.
- W3209504420 hasConcept C2781238097 @default.
- W3209504420 hasConcept C2987834672 @default.
- W3209504420 hasConcept C31972630 @default.
- W3209504420 hasConcept C37914503 @default.
- W3209504420 hasConcept C41008148 @default.
- W3209504420 hasConcept C62520636 @default.
- W3209504420 hasConcept C63584917 @default.
- W3209504420 hasConcept C64876066 @default.
- W3209504420 hasConceptScore W3209504420C119599485 @default.
- W3209504420 hasConceptScore W3209504420C119857082 @default.
- W3209504420 hasConceptScore W3209504420C120665830 @default.
- W3209504420 hasConceptScore W3209504420C121332964 @default.
- W3209504420 hasConceptScore W3209504420C127413603 @default.