Matches in SemOpenAlex for { <https://semopenalex.org/work/W4366308342> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4366308342 abstract "Action classification has made great progress, but segmenting and recognizing actions from long videos remains a challenging problem. Recently, Transformer-based models with strong sequence modeling ability have succeeded in many se-quence modeling tasks. However, the lack of inductive bias and the difficulty of handling long video sequences limit the application of the Transformer in the action segmentation task. In order to explore the potential of the Transformer in this task, we replace some specific linear layers in the vanilla Transformer with dilated temporal convolution, and a sparse attention mechanism is utilized to reduce the time and space complexities to process long video sequences. Besides, directly using frame-wise classification loss to train the model will cause that frames at boundaries of actions are treated equally with those in the middle of actions, and the learned features are not sensitive to boundaries. We propose a new local log-context attention module to predict whether each frame is at the beginning, middle, or end of an action. Since boundary frames are similar to their neighboring frames of different classes, our similarity-based boundary prediction helps learn more discriminative features. Extensive experiments on three datasets show the effectiveness of our method." @default.
- W4366308342 created "2023-04-20" @default.
- W4366308342 creator A5003513450 @default.
- W4366308342 creator A5011937977 @default.
- W4366308342 creator A5015095434 @default.
- W4366308342 creator A5018073672 @default.
- W4366308342 creator A5027552992 @default.
- W4366308342 creator A5088475144 @default.
- W4366308342 date "2022-10-01" @default.
- W4366308342 modified "2023-09-27" @default.
- W4366308342 title "Convolutional Transformer with Similarity-based Boundary Prediction for Action Segmentation" @default.
- W4366308342 cites W2031688197 @default.
- W4366308342 cites W2099614498 @default.
- W4366308342 cites W2109698606 @default.
- W4366308342 cites W2461621749 @default.
- W4366308342 cites W2550143307 @default.
- W4366308342 cites W2799262584 @default.
- W4366308342 cites W2963524571 @default.
- W4366308342 cites W2963853051 @default.
- W4366308342 cites W3034802267 @default.
- W4366308342 cites W3083550439 @default.
- W4366308342 cites W3108772932 @default.
- W4366308342 cites W3119038403 @default.
- W4366308342 cites W3210455685 @default.
- W4366308342 doi "https://doi.org/10.1109/ictai56018.2022.00131" @default.
- W4366308342 hasPublicationYear "2022" @default.
- W4366308342 type Work @default.
- W4366308342 citedByCount "0" @default.
- W4366308342 crossrefType "proceedings-article" @default.
- W4366308342 hasAuthorship W4366308342A5003513450 @default.
- W4366308342 hasAuthorship W4366308342A5011937977 @default.
- W4366308342 hasAuthorship W4366308342A5015095434 @default.
- W4366308342 hasAuthorship W4366308342A5018073672 @default.
- W4366308342 hasAuthorship W4366308342A5027552992 @default.
- W4366308342 hasAuthorship W4366308342A5088475144 @default.
- W4366308342 hasConcept C119599485 @default.
- W4366308342 hasConcept C119857082 @default.
- W4366308342 hasConcept C127413603 @default.
- W4366308342 hasConcept C153180895 @default.
- W4366308342 hasConcept C154945302 @default.
- W4366308342 hasConcept C165801399 @default.
- W4366308342 hasConcept C31972630 @default.
- W4366308342 hasConcept C41008148 @default.
- W4366308342 hasConcept C66322947 @default.
- W4366308342 hasConcept C89600930 @default.
- W4366308342 hasConcept C97931131 @default.
- W4366308342 hasConceptScore W4366308342C119599485 @default.
- W4366308342 hasConceptScore W4366308342C119857082 @default.
- W4366308342 hasConceptScore W4366308342C127413603 @default.
- W4366308342 hasConceptScore W4366308342C153180895 @default.
- W4366308342 hasConceptScore W4366308342C154945302 @default.
- W4366308342 hasConceptScore W4366308342C165801399 @default.
- W4366308342 hasConceptScore W4366308342C31972630 @default.
- W4366308342 hasConceptScore W4366308342C41008148 @default.
- W4366308342 hasConceptScore W4366308342C66322947 @default.
- W4366308342 hasConceptScore W4366308342C89600930 @default.
- W4366308342 hasConceptScore W4366308342C97931131 @default.
- W4366308342 hasLocation W43663083421 @default.
- W4366308342 hasOpenAccess W4366308342 @default.
- W4366308342 hasPrimaryLocation W43663083421 @default.
- W4366308342 hasRelatedWork W1652783584 @default.
- W4366308342 hasRelatedWork W1669643531 @default.
- W4366308342 hasRelatedWork W1990254706 @default.
- W4366308342 hasRelatedWork W2024160000 @default.
- W4366308342 hasRelatedWork W2404514746 @default.
- W4366308342 hasRelatedWork W2510758617 @default.
- W4366308342 hasRelatedWork W2517104666 @default.
- W4366308342 hasRelatedWork W2773500201 @default.
- W4366308342 hasRelatedWork W4287995534 @default.
- W4366308342 hasRelatedWork W2073139667 @default.
- W4366308342 isParatext "false" @default.
- W4366308342 isRetracted "false" @default.
- W4366308342 workType "article" @default.