Matches in SemOpenAlex for { <https://semopenalex.org/work/W4286751018> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4286751018 abstract "Clustering is a ubiquitous tool in unsupervised learning. Most of the existing self-supervised representation learning methods typically cluster samples based on visually dominant features. While this works well for image-based self-supervision, it often fails for videos, which require understanding motion rather than focusing on background. Using optical flow as complementary information to RGB can alleviate this problem. However, we observe that a naive combination of the two views does not provide meaningful gains. In this paper, we propose a principled way to combine two views. Specifically, we propose a novel clustering strategy where we use the initial cluster assignment of each view as prior to guide the final cluster assignment of the other view. This idea will enforce similar cluster structures for both views, and the formed clusters will be semantically abstract and robust to noisy inputs coming from each individual view. Additionally, we propose a novel regularization strategy to address the feature collapse problem, which is common in cluster-based self-supervised learning methods. Our extensive evaluation shows the effectiveness of our learned representations on downstream tasks, e.g., video retrieval and action recognition. Specifically, we outperform the state of the art by 7% on UCF and 4% on HMDB for video retrieval, and 5% on UCF and 6% on HMDB for video classification" @default.
- W4286751018 created "2022-07-23" @default.
- W4286751018 creator A5013038112 @default.
- W4286751018 creator A5021016566 @default.
- W4286751018 creator A5041092666 @default.
- W4286751018 creator A5052579251 @default.
- W4286751018 creator A5073512409 @default.
- W4286751018 date "2022-07-20" @default.
- W4286751018 modified "2023-10-16" @default.
- W4286751018 title "GOCA: Guided Online Cluster Assignment for Self-Supervised Video Representation Learning" @default.
- W4286751018 doi "https://doi.org/10.48550/arxiv.2207.10158" @default.
- W4286751018 hasPublicationYear "2022" @default.
- W4286751018 type Work @default.
- W4286751018 citedByCount "0" @default.
- W4286751018 crossrefType "posted-content" @default.
- W4286751018 hasAuthorship W4286751018A5013038112 @default.
- W4286751018 hasAuthorship W4286751018A5021016566 @default.
- W4286751018 hasAuthorship W4286751018A5041092666 @default.
- W4286751018 hasAuthorship W4286751018A5052579251 @default.
- W4286751018 hasAuthorship W4286751018A5073512409 @default.
- W4286751018 hasBestOaLocation W42867510181 @default.
- W4286751018 hasConcept C115961682 @default.
- W4286751018 hasConcept C119857082 @default.
- W4286751018 hasConcept C138885662 @default.
- W4286751018 hasConcept C153180895 @default.
- W4286751018 hasConcept C154945302 @default.
- W4286751018 hasConcept C155542232 @default.
- W4286751018 hasConcept C17744445 @default.
- W4286751018 hasConcept C199539241 @default.
- W4286751018 hasConcept C2776135515 @default.
- W4286751018 hasConcept C2776359362 @default.
- W4286751018 hasConcept C2776401178 @default.
- W4286751018 hasConcept C41008148 @default.
- W4286751018 hasConcept C41895202 @default.
- W4286751018 hasConcept C59404180 @default.
- W4286751018 hasConcept C73555534 @default.
- W4286751018 hasConcept C8038995 @default.
- W4286751018 hasConcept C82990744 @default.
- W4286751018 hasConcept C94625758 @default.
- W4286751018 hasConceptScore W4286751018C115961682 @default.
- W4286751018 hasConceptScore W4286751018C119857082 @default.
- W4286751018 hasConceptScore W4286751018C138885662 @default.
- W4286751018 hasConceptScore W4286751018C153180895 @default.
- W4286751018 hasConceptScore W4286751018C154945302 @default.
- W4286751018 hasConceptScore W4286751018C155542232 @default.
- W4286751018 hasConceptScore W4286751018C17744445 @default.
- W4286751018 hasConceptScore W4286751018C199539241 @default.
- W4286751018 hasConceptScore W4286751018C2776135515 @default.
- W4286751018 hasConceptScore W4286751018C2776359362 @default.
- W4286751018 hasConceptScore W4286751018C2776401178 @default.
- W4286751018 hasConceptScore W4286751018C41008148 @default.
- W4286751018 hasConceptScore W4286751018C41895202 @default.
- W4286751018 hasConceptScore W4286751018C59404180 @default.
- W4286751018 hasConceptScore W4286751018C73555534 @default.
- W4286751018 hasConceptScore W4286751018C8038995 @default.
- W4286751018 hasConceptScore W4286751018C82990744 @default.
- W4286751018 hasConceptScore W4286751018C94625758 @default.
- W4286751018 hasLocation W42867510181 @default.
- W4286751018 hasOpenAccess W4286751018 @default.
- W4286751018 hasPrimaryLocation W42867510181 @default.
- W4286751018 hasRelatedWork W2695951553 @default.
- W4286751018 hasRelatedWork W2775464024 @default.
- W4286751018 hasRelatedWork W2891219753 @default.
- W4286751018 hasRelatedWork W2902482704 @default.
- W4286751018 hasRelatedWork W2908875379 @default.
- W4286751018 hasRelatedWork W2922225723 @default.
- W4286751018 hasRelatedWork W2966870361 @default.
- W4286751018 hasRelatedWork W2970216048 @default.
- W4286751018 hasRelatedWork W3093564170 @default.
- W4286751018 hasRelatedWork W4221136938 @default.
- W4286751018 isParatext "false" @default.
- W4286751018 isRetracted "false" @default.
- W4286751018 workType "article" @default.