Matches in SemOpenAlex for { <https://semopenalex.org/work/W2894724248> ?p ?o ?g. }
- W2894724248 abstract "In this paper, we propose a convolutional layer inspired by optical flow algorithms to learn motion representations. Our representation flow layer is a fully-differentiable layer designed to capture the `flow' of any representation channel within a convolutional neural network for action recognition. Its parameters for iterative flow optimization are learned in an end-to-end fashion together with the other CNN model parameters, maximizing the action recognition performance. Furthermore, we newly introduce the concept of learning `flow of flow' representations by stacking multiple representation flow layers. We conducted extensive experimental evaluations, confirming its advantages over previous recognition models using traditional optical flows in both computational speed and performance. Code/models available here: this https URL" @default.
- W2894724248 created "2018-10-12" @default.
- W2894724248 creator A5084829008 @default.
- W2894724248 creator A5086527941 @default.
- W2894724248 date "2018-10-02" @default.
- W2894724248 modified "2023-09-27" @default.
- W2894724248 title "Representation Flow for Action Recognition" @default.
- W2894724248 cites W1578985305 @default.
- W2894724248 cites W1867429401 @default.
- W2894724248 cites W1923404803 @default.
- W2894724248 cites W2016053056 @default.
- W2894724248 cites W2059612594 @default.
- W2894724248 cites W2126574503 @default.
- W2894724248 cites W2126579184 @default.
- W2894724248 cites W2156303437 @default.
- W2894724248 cites W2183341477 @default.
- W2894724248 cites W2194775991 @default.
- W2894724248 cites W2308045930 @default.
- W2894724248 cites W2342662179 @default.
- W2894724248 cites W2462996230 @default.
- W2894724248 cites W2553594924 @default.
- W2894724248 cites W2566769621 @default.
- W2894724248 cites W2619947201 @default.
- W2894724248 cites W2773514261 @default.
- W2894724248 cites W2799261915 @default.
- W2894724248 cites W2884797191 @default.
- W2894724248 cites W2963091558 @default.
- W2894724248 cites W2963155035 @default.
- W2894724248 cites W2963216700 @default.
- W2894724248 cites W2963484593 @default.
- W2894724248 cites W2963524571 @default.
- W2894724248 cites W2963775820 @default.
- W2894724248 cites W2964094092 @default.
- W2894724248 cites W2964156315 @default.
- W2894724248 cites W2964297311 @default.
- W2894724248 hasPublicationYear "2018" @default.
- W2894724248 type Work @default.
- W2894724248 sameAs 2894724248 @default.
- W2894724248 citedByCount "2" @default.
- W2894724248 countsByYear W28947242482018 @default.
- W2894724248 countsByYear W28947242482021 @default.
- W2894724248 crossrefType "posted-content" @default.
- W2894724248 hasAuthorship W2894724248A5084829008 @default.
- W2894724248 hasAuthorship W2894724248A5086527941 @default.
- W2894724248 hasConcept C115961682 @default.
- W2894724248 hasConcept C121332964 @default.
- W2894724248 hasConcept C153180895 @default.
- W2894724248 hasConcept C154945302 @default.
- W2894724248 hasConcept C155542232 @default.
- W2894724248 hasConcept C177264268 @default.
- W2894724248 hasConcept C17744445 @default.
- W2894724248 hasConcept C178790620 @default.
- W2894724248 hasConcept C185592680 @default.
- W2894724248 hasConcept C199360897 @default.
- W2894724248 hasConcept C199539241 @default.
- W2894724248 hasConcept C2524010 @default.
- W2894724248 hasConcept C2776359362 @default.
- W2894724248 hasConcept C2776760102 @default.
- W2894724248 hasConcept C2777212361 @default.
- W2894724248 hasConcept C2779227376 @default.
- W2894724248 hasConcept C2780791683 @default.
- W2894724248 hasConcept C2987834672 @default.
- W2894724248 hasConcept C33347731 @default.
- W2894724248 hasConcept C33923547 @default.
- W2894724248 hasConcept C38349280 @default.
- W2894724248 hasConcept C41008148 @default.
- W2894724248 hasConcept C46141821 @default.
- W2894724248 hasConcept C62520636 @default.
- W2894724248 hasConcept C80444323 @default.
- W2894724248 hasConcept C81363708 @default.
- W2894724248 hasConcept C94625758 @default.
- W2894724248 hasConceptScore W2894724248C115961682 @default.
- W2894724248 hasConceptScore W2894724248C121332964 @default.
- W2894724248 hasConceptScore W2894724248C153180895 @default.
- W2894724248 hasConceptScore W2894724248C154945302 @default.
- W2894724248 hasConceptScore W2894724248C155542232 @default.
- W2894724248 hasConceptScore W2894724248C177264268 @default.
- W2894724248 hasConceptScore W2894724248C17744445 @default.
- W2894724248 hasConceptScore W2894724248C178790620 @default.
- W2894724248 hasConceptScore W2894724248C185592680 @default.
- W2894724248 hasConceptScore W2894724248C199360897 @default.
- W2894724248 hasConceptScore W2894724248C199539241 @default.
- W2894724248 hasConceptScore W2894724248C2524010 @default.
- W2894724248 hasConceptScore W2894724248C2776359362 @default.
- W2894724248 hasConceptScore W2894724248C2776760102 @default.
- W2894724248 hasConceptScore W2894724248C2777212361 @default.
- W2894724248 hasConceptScore W2894724248C2779227376 @default.
- W2894724248 hasConceptScore W2894724248C2780791683 @default.
- W2894724248 hasConceptScore W2894724248C2987834672 @default.
- W2894724248 hasConceptScore W2894724248C33347731 @default.
- W2894724248 hasConceptScore W2894724248C33923547 @default.
- W2894724248 hasConceptScore W2894724248C38349280 @default.
- W2894724248 hasConceptScore W2894724248C41008148 @default.
- W2894724248 hasConceptScore W2894724248C46141821 @default.
- W2894724248 hasConceptScore W2894724248C62520636 @default.
- W2894724248 hasConceptScore W2894724248C80444323 @default.
- W2894724248 hasConceptScore W2894724248C81363708 @default.
- W2894724248 hasConceptScore W2894724248C94625758 @default.
- W2894724248 hasLocation W28947242481 @default.
- W2894724248 hasOpenAccess W2894724248 @default.