Matches in SemOpenAlex for { <https://semopenalex.org/work/W3201942032> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W3201942032 abstract "This paper studies the problem of learning self-supervised representations on videos. In contrast to image modality that only requires appearance information on objects or scenes, video needs to further explore the relations between multiple frames/clips along the temporal dimension. However, the recent proposed contrastive-based self-supervised frameworks do not grasp such relations explicitly since they simply utilize two augmented clips from the same video and compare their distance without referring to their temporal relation. To address this, we present a contrast-and-order representation (CORP) framework for learning self-supervised video representations that can automatically capture both the appearance information within each frame and temporal information across different frames. In particular, given two video clips, our model first predicts whether they come from the same input video, and then predict the temporal ordering of the clips if they come from the same video. We also propose a novel decoupling attention method to learn symmetric similarity (contrast) and anti-symmetric patterns (order). Such design involves neither extra parameters nor computation, but can speed up the learning process and improve accuracy compared to the vanilla multi-head attention. We extensively validate the representation ability of our learned video features for the downstream action recognition task on Kinetics-400 and Something-something V2. Our method outperforms previous state-of-the-arts by a significant margin." @default.
- W3201942032 created "2021-10-11" @default.
- W3201942032 creator A5015367334 @default.
- W3201942032 creator A5016771874 @default.
- W3201942032 creator A5048675276 @default.
- W3201942032 creator A5057959136 @default.
- W3201942032 creator A5069339294 @default.
- W3201942032 creator A5072350518 @default.
- W3201942032 date "2021-10-01" @default.
- W3201942032 modified "2023-10-16" @default.
- W3201942032 title "Contrast and Order Representations for Video Self-supervised Learning" @default.
- W3201942032 cites W1903029394 @default.
- W3201942032 cites W2126579184 @default.
- W3201942032 cites W2550462002 @default.
- W3201942032 cites W2625366777 @default.
- W3201942032 cites W2798271879 @default.
- W3201942032 cites W2927438889 @default.
- W3201942032 cites W2948242301 @default.
- W3201942032 cites W2963091558 @default.
- W3201942032 cites W2963150697 @default.
- W3201942032 cites W2963466847 @default.
- W3201942032 cites W2963524571 @default.
- W3201942032 cites W2963631366 @default.
- W3201942032 cites W2963645879 @default.
- W3201942032 cites W2963814513 @default.
- W3201942032 cites W2963855133 @default.
- W3201942032 cites W2963863119 @default.
- W3201942032 cites W2964037671 @default.
- W3201942032 cites W2964700958 @default.
- W3201942032 cites W2969985801 @default.
- W3201942032 cites W2981851019 @default.
- W3201942032 cites W2990152177 @default.
- W3201942032 cites W3034215340 @default.
- W3201942032 cites W3035524453 @default.
- W3201942032 cites W3035635319 @default.
- W3201942032 cites W3170837227 @default.
- W3201942032 cites W343636949 @default.
- W3201942032 doi "https://doi.org/10.1109/iccv48922.2021.00784" @default.
- W3201942032 hasPublicationYear "2021" @default.
- W3201942032 type Work @default.
- W3201942032 sameAs 3201942032 @default.
- W3201942032 citedByCount "19" @default.
- W3201942032 countsByYear W32019420322021 @default.
- W3201942032 countsByYear W32019420322022 @default.
- W3201942032 countsByYear W32019420322023 @default.
- W3201942032 crossrefType "proceedings-article" @default.
- W3201942032 hasAuthorship W3201942032A5015367334 @default.
- W3201942032 hasAuthorship W3201942032A5016771874 @default.
- W3201942032 hasAuthorship W3201942032A5048675276 @default.
- W3201942032 hasAuthorship W3201942032A5057959136 @default.
- W3201942032 hasAuthorship W3201942032A5069339294 @default.
- W3201942032 hasAuthorship W3201942032A5072350518 @default.
- W3201942032 hasConcept C124101348 @default.
- W3201942032 hasConcept C153180895 @default.
- W3201942032 hasConcept C154945302 @default.
- W3201942032 hasConcept C171268870 @default.
- W3201942032 hasConcept C17744445 @default.
- W3201942032 hasConcept C199360897 @default.
- W3201942032 hasConcept C199539241 @default.
- W3201942032 hasConcept C25343380 @default.
- W3201942032 hasConcept C2776359362 @default.
- W3201942032 hasConcept C2776502983 @default.
- W3201942032 hasConcept C2778739407 @default.
- W3201942032 hasConcept C31972630 @default.
- W3201942032 hasConcept C41008148 @default.
- W3201942032 hasConcept C94625758 @default.
- W3201942032 hasConceptScore W3201942032C124101348 @default.
- W3201942032 hasConceptScore W3201942032C153180895 @default.
- W3201942032 hasConceptScore W3201942032C154945302 @default.
- W3201942032 hasConceptScore W3201942032C171268870 @default.
- W3201942032 hasConceptScore W3201942032C17744445 @default.
- W3201942032 hasConceptScore W3201942032C199360897 @default.
- W3201942032 hasConceptScore W3201942032C199539241 @default.
- W3201942032 hasConceptScore W3201942032C25343380 @default.
- W3201942032 hasConceptScore W3201942032C2776359362 @default.
- W3201942032 hasConceptScore W3201942032C2776502983 @default.
- W3201942032 hasConceptScore W3201942032C2778739407 @default.
- W3201942032 hasConceptScore W3201942032C31972630 @default.
- W3201942032 hasConceptScore W3201942032C41008148 @default.
- W3201942032 hasConceptScore W3201942032C94625758 @default.
- W3201942032 hasLocation W32019420321 @default.
- W3201942032 hasOpenAccess W3201942032 @default.
- W3201942032 hasPrimaryLocation W32019420321 @default.
- W3201942032 hasRelatedWork W1910101490 @default.
- W3201942032 hasRelatedWork W1987421842 @default.
- W3201942032 hasRelatedWork W2389377526 @default.
- W3201942032 hasRelatedWork W2557924869 @default.
- W3201942032 hasRelatedWork W2888481972 @default.
- W3201942032 hasRelatedWork W3185561939 @default.
- W3201942032 hasRelatedWork W3192511943 @default.
- W3201942032 hasRelatedWork W4200303201 @default.
- W3201942032 hasRelatedWork W4321376912 @default.
- W3201942032 hasRelatedWork W4386794497 @default.
- W3201942032 isParatext "false" @default.
- W3201942032 isRetracted "false" @default.
- W3201942032 magId "3201942032" @default.
- W3201942032 workType "article" @default.