Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313305652> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W4313305652 abstract "Video representation learning has been successful in video-text pre-training for zero-shot transfer, where each sentence is trained to be close to the paired video clips in a common feature space. For long videos, given a paragraph of description where the sentences describe different segments of the video, by matching all sentence-clip pairs, the paragraph and the full video are aligned implicitly. However, such unit-level similarity measure may ignore the global temporal context over a long time span, which inevitably limits the generalization ability. In this paper, we propose a contrastive learning framework TempCLR to compare the full video and the paragraph explicitly. As the video/paragraph is formulated as a sequence of clips/sentences, under the constraint of their temporal order, we use dynamic time warping to compute the minimum cumulative cost over sentence-clip pairs as the sequence-level distance. To explore the temporal dynamics, we break the consistency of temporal order by shuffling the video clips or sentences according to the temporal granularity. In this way, we obtain the representations for clips/sentences, which perceive the temporal information and thus facilitate the sequence alignment. In addition to pre-training on the video and paragraph, our approach can also generalize on the matching between different video instances. We evaluate our approach on video retrieval, action step localization, and few-shot action recognition, and achieve consistent performance gain over all three tasks. Detailed ablation studies are provided to justify the approach design." @default.
- W4313305652 created "2023-01-06" @default.
- W4313305652 creator A5012324763 @default.
- W4313305652 creator A5012558857 @default.
- W4313305652 creator A5017578261 @default.
- W4313305652 creator A5025159157 @default.
- W4313305652 creator A5037340457 @default.
- W4313305652 creator A5056500637 @default.
- W4313305652 creator A5071372959 @default.
- W4313305652 date "2022-12-28" @default.
- W4313305652 modified "2023-10-18" @default.
- W4313305652 title "TempCLR: Temporal Alignment Representation with Contrastive Learning" @default.
- W4313305652 doi "https://doi.org/10.48550/arxiv.2212.13738" @default.
- W4313305652 hasPublicationYear "2022" @default.
- W4313305652 type Work @default.
- W4313305652 citedByCount "0" @default.
- W4313305652 crossrefType "posted-content" @default.
- W4313305652 hasAuthorship W4313305652A5012324763 @default.
- W4313305652 hasAuthorship W4313305652A5012558857 @default.
- W4313305652 hasAuthorship W4313305652A5017578261 @default.
- W4313305652 hasAuthorship W4313305652A5025159157 @default.
- W4313305652 hasAuthorship W4313305652A5037340457 @default.
- W4313305652 hasAuthorship W4313305652A5056500637 @default.
- W4313305652 hasAuthorship W4313305652A5071372959 @default.
- W4313305652 hasBestOaLocation W43133056521 @default.
- W4313305652 hasConcept C103278499 @default.
- W4313305652 hasConcept C105795698 @default.
- W4313305652 hasConcept C115961682 @default.
- W4313305652 hasConcept C136764020 @default.
- W4313305652 hasConcept C138885662 @default.
- W4313305652 hasConcept C151730666 @default.
- W4313305652 hasConcept C154945302 @default.
- W4313305652 hasConcept C165064840 @default.
- W4313305652 hasConcept C204321447 @default.
- W4313305652 hasConcept C2776401178 @default.
- W4313305652 hasConcept C2777206241 @default.
- W4313305652 hasConcept C2777530160 @default.
- W4313305652 hasConcept C2778739407 @default.
- W4313305652 hasConcept C2779343474 @default.
- W4313305652 hasConcept C28490314 @default.
- W4313305652 hasConcept C33923547 @default.
- W4313305652 hasConcept C41008148 @default.
- W4313305652 hasConcept C41895202 @default.
- W4313305652 hasConcept C86803240 @default.
- W4313305652 hasConcept C88516994 @default.
- W4313305652 hasConceptScore W4313305652C103278499 @default.
- W4313305652 hasConceptScore W4313305652C105795698 @default.
- W4313305652 hasConceptScore W4313305652C115961682 @default.
- W4313305652 hasConceptScore W4313305652C136764020 @default.
- W4313305652 hasConceptScore W4313305652C138885662 @default.
- W4313305652 hasConceptScore W4313305652C151730666 @default.
- W4313305652 hasConceptScore W4313305652C154945302 @default.
- W4313305652 hasConceptScore W4313305652C165064840 @default.
- W4313305652 hasConceptScore W4313305652C204321447 @default.
- W4313305652 hasConceptScore W4313305652C2776401178 @default.
- W4313305652 hasConceptScore W4313305652C2777206241 @default.
- W4313305652 hasConceptScore W4313305652C2777530160 @default.
- W4313305652 hasConceptScore W4313305652C2778739407 @default.
- W4313305652 hasConceptScore W4313305652C2779343474 @default.
- W4313305652 hasConceptScore W4313305652C28490314 @default.
- W4313305652 hasConceptScore W4313305652C33923547 @default.
- W4313305652 hasConceptScore W4313305652C41008148 @default.
- W4313305652 hasConceptScore W4313305652C41895202 @default.
- W4313305652 hasConceptScore W4313305652C86803240 @default.
- W4313305652 hasConceptScore W4313305652C88516994 @default.
- W4313305652 hasLocation W43133056521 @default.
- W4313305652 hasOpenAccess W4313305652 @default.
- W4313305652 hasPrimaryLocation W43133056521 @default.
- W4313305652 hasRelatedWork W1519302135 @default.
- W4313305652 hasRelatedWork W1584662895 @default.
- W4313305652 hasRelatedWork W1601017932 @default.
- W4313305652 hasRelatedWork W2013781749 @default.
- W4313305652 hasRelatedWork W2070147364 @default.
- W4313305652 hasRelatedWork W2116869842 @default.
- W4313305652 hasRelatedWork W2168158993 @default.
- W4313305652 hasRelatedWork W2295253065 @default.
- W4313305652 hasRelatedWork W2357163346 @default.
- W4313305652 hasRelatedWork W2379635239 @default.
- W4313305652 isParatext "false" @default.
- W4313305652 isRetracted "false" @default.
- W4313305652 workType "article" @default.