Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386075563> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W4386075563 abstract "Retrieval augmented models are becoming increasingly popular for computer vision tasks after their recent success in NLP problems. The goal is to enhance the recognition capabilities of the model by retrieving similar examples for the visual input from an external memory set. In this work, we introduce an attention-based memory module, which learns the importance of each retrieved example from the memory. Compared to existing approaches, our method removes the influence of the irrelevant retrieved examples, and retains those that are beneficial to the input query. We also thoroughly study various ways of constructing the memory dataset. Our experiments show the benefit of using a massive-scale memory dataset of 1B image-text pairs, and demonstrate the performance of different memory representations. We evaluate our method in three different classification tasks, namely long-tailed recognition, learning with noisy labels, and fine-grained classification, and show that it achieves state-of-the-art accuracies in ImageNet-LT, Places-LT and Webvision datasets." @default.
- W4386075563 created "2023-08-23" @default.
- W4386075563 creator A5045217258 @default.
- W4386075563 creator A5063473743 @default.
- W4386075563 creator A5088574130 @default.
- W4386075563 date "2023-06-01" @default.
- W4386075563 modified "2023-10-11" @default.
- W4386075563 title "Improving Image Recognition by Retrieving from Web-Scale Image-Text Data" @default.
- W4386075563 cites W2023954349 @default.
- W4386075563 cites W2108598243 @default.
- W4386075563 cites W2117539524 @default.
- W4386075563 cites W2183341477 @default.
- W4386075563 cites W2250384498 @default.
- W4386075563 cites W2440599146 @default.
- W4386075563 cites W2962933664 @default.
- W4386075563 cites W3034601242 @default.
- W4386075563 cites W3172942063 @default.
- W4386075563 cites W3173685130 @default.
- W4386075563 cites W3177200443 @default.
- W4386075563 cites W3177357839 @default.
- W4386075563 cites W3182707920 @default.
- W4386075563 cites W3213454282 @default.
- W4386075563 cites W4214718285 @default.
- W4386075563 cites W4285122897 @default.
- W4386075563 cites W4312305885 @default.
- W4386075563 cites W4312910992 @default.
- W4386075563 cites W4312920106 @default.
- W4386075563 cites W4313002532 @default.
- W4386075563 doi "https://doi.org/10.1109/cvpr52729.2023.01849" @default.
- W4386075563 hasPublicationYear "2023" @default.
- W4386075563 type Work @default.
- W4386075563 citedByCount "1" @default.
- W4386075563 crossrefType "proceedings-article" @default.
- W4386075563 hasAuthorship W4386075563A5045217258 @default.
- W4386075563 hasAuthorship W4386075563A5063473743 @default.
- W4386075563 hasAuthorship W4386075563A5088574130 @default.
- W4386075563 hasConcept C115961682 @default.
- W4386075563 hasConcept C119857082 @default.
- W4386075563 hasConcept C121332964 @default.
- W4386075563 hasConcept C153180895 @default.
- W4386075563 hasConcept C154945302 @default.
- W4386075563 hasConcept C1667742 @default.
- W4386075563 hasConcept C177264268 @default.
- W4386075563 hasConcept C199360897 @default.
- W4386075563 hasConcept C23123220 @default.
- W4386075563 hasConcept C2778755073 @default.
- W4386075563 hasConcept C41008148 @default.
- W4386075563 hasConcept C62520636 @default.
- W4386075563 hasConceptScore W4386075563C115961682 @default.
- W4386075563 hasConceptScore W4386075563C119857082 @default.
- W4386075563 hasConceptScore W4386075563C121332964 @default.
- W4386075563 hasConceptScore W4386075563C153180895 @default.
- W4386075563 hasConceptScore W4386075563C154945302 @default.
- W4386075563 hasConceptScore W4386075563C1667742 @default.
- W4386075563 hasConceptScore W4386075563C177264268 @default.
- W4386075563 hasConceptScore W4386075563C199360897 @default.
- W4386075563 hasConceptScore W4386075563C23123220 @default.
- W4386075563 hasConceptScore W4386075563C2778755073 @default.
- W4386075563 hasConceptScore W4386075563C41008148 @default.
- W4386075563 hasConceptScore W4386075563C62520636 @default.
- W4386075563 hasLocation W43860755631 @default.
- W4386075563 hasOpenAccess W4386075563 @default.
- W4386075563 hasPrimaryLocation W43860755631 @default.
- W4386075563 hasRelatedWork W1926504679 @default.
- W4386075563 hasRelatedWork W2027614354 @default.
- W4386075563 hasRelatedWork W2060690676 @default.
- W4386075563 hasRelatedWork W2088097596 @default.
- W4386075563 hasRelatedWork W2353314619 @default.
- W4386075563 hasRelatedWork W2519881370 @default.
- W4386075563 hasRelatedWork W2532409108 @default.
- W4386075563 hasRelatedWork W3097853387 @default.
- W4386075563 hasRelatedWork W3150772620 @default.
- W4386075563 hasRelatedWork W4287330208 @default.
- W4386075563 isParatext "false" @default.
- W4386075563 isRetracted "false" @default.
- W4386075563 workType "article" @default.