Matches in SemOpenAlex for { <https://semopenalex.org/work/W4382466690> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4382466690 endingPage "763" @default.
- W4382466690 startingPage "755" @default.
- W4382466690 abstract "As few-shot object detectors are often trained with abundant base samples and fine-tuned on few-shot novel examples, the learned models are usually biased to base classes and sensitive to the variance of novel examples. To address this issue, we propose a meta-learning framework with two novel feature aggregation schemes. More precisely, we first present a Class-Agnostic Aggregation (CAA) method, where the query and support features can be aggregated regardless of their categories. The interactions between different classes encourage class-agnostic representations and reduce confusion between base and novel classes. Based on the CAA, we then propose a Variational Feature Aggregation (VFA) method, which encodes support examples into class-level support features for robust feature aggregation. We use a variational autoencoder to estimate class distributions and sample variational features from distributions that are more robust to the variance of support examples. Besides, we decouple classification and regression tasks so that VFA is performed on the classification branch without affecting object localization. Extensive experiments on PASCAL VOC and COCO demonstrate that our method significantly outperforms a strong baseline (up to 16%) and previous state-of-the-art methods (4% in average)." @default.
- W4382466690 created "2023-06-29" @default.
- W4382466690 creator A5001680256 @default.
- W4382466690 creator A5005276003 @default.
- W4382466690 creator A5026379418 @default.
- W4382466690 creator A5070798635 @default.
- W4382466690 creator A5073032922 @default.
- W4382466690 date "2023-06-26" @default.
- W4382466690 modified "2023-09-23" @default.
- W4382466690 title "Few-Shot Object Detection via Variational Feature Aggregation" @default.
- W4382466690 doi "https://doi.org/10.1609/aaai.v37i1.25153" @default.
- W4382466690 hasPublicationYear "2023" @default.
- W4382466690 type Work @default.
- W4382466690 citedByCount "0" @default.
- W4382466690 crossrefType "journal-article" @default.
- W4382466690 hasAuthorship W4382466690A5001680256 @default.
- W4382466690 hasAuthorship W4382466690A5005276003 @default.
- W4382466690 hasAuthorship W4382466690A5026379418 @default.
- W4382466690 hasAuthorship W4382466690A5070798635 @default.
- W4382466690 hasAuthorship W4382466690A5073032922 @default.
- W4382466690 hasBestOaLocation W43824666901 @default.
- W4382466690 hasConcept C101738243 @default.
- W4382466690 hasConcept C108583219 @default.
- W4382466690 hasConcept C121955636 @default.
- W4382466690 hasConcept C138885662 @default.
- W4382466690 hasConcept C144133560 @default.
- W4382466690 hasConcept C153180895 @default.
- W4382466690 hasConcept C154945302 @default.
- W4382466690 hasConcept C196083921 @default.
- W4382466690 hasConcept C199360897 @default.
- W4382466690 hasConcept C2776401178 @default.
- W4382466690 hasConcept C2777212361 @default.
- W4382466690 hasConcept C2781238097 @default.
- W4382466690 hasConcept C41008148 @default.
- W4382466690 hasConcept C41895202 @default.
- W4382466690 hasConcept C75608658 @default.
- W4382466690 hasConcept C95623464 @default.
- W4382466690 hasConceptScore W4382466690C101738243 @default.
- W4382466690 hasConceptScore W4382466690C108583219 @default.
- W4382466690 hasConceptScore W4382466690C121955636 @default.
- W4382466690 hasConceptScore W4382466690C138885662 @default.
- W4382466690 hasConceptScore W4382466690C144133560 @default.
- W4382466690 hasConceptScore W4382466690C153180895 @default.
- W4382466690 hasConceptScore W4382466690C154945302 @default.
- W4382466690 hasConceptScore W4382466690C196083921 @default.
- W4382466690 hasConceptScore W4382466690C199360897 @default.
- W4382466690 hasConceptScore W4382466690C2776401178 @default.
- W4382466690 hasConceptScore W4382466690C2777212361 @default.
- W4382466690 hasConceptScore W4382466690C2781238097 @default.
- W4382466690 hasConceptScore W4382466690C41008148 @default.
- W4382466690 hasConceptScore W4382466690C41895202 @default.
- W4382466690 hasConceptScore W4382466690C75608658 @default.
- W4382466690 hasConceptScore W4382466690C95623464 @default.
- W4382466690 hasIssue "1" @default.
- W4382466690 hasLocation W43824666901 @default.
- W4382466690 hasOpenAccess W4382466690 @default.
- W4382466690 hasPrimaryLocation W43824666901 @default.
- W4382466690 hasRelatedWork W2292254049 @default.
- W4382466690 hasRelatedWork W2572600474 @default.
- W4382466690 hasRelatedWork W2776466379 @default.
- W4382466690 hasRelatedWork W2785535669 @default.
- W4382466690 hasRelatedWork W2897995864 @default.
- W4382466690 hasRelatedWork W2998168123 @default.
- W4382466690 hasRelatedWork W3133754756 @default.
- W4382466690 hasRelatedWork W4220775285 @default.
- W4382466690 hasRelatedWork W4221015625 @default.
- W4382466690 hasRelatedWork W4287995534 @default.
- W4382466690 hasVolume "37" @default.
- W4382466690 isParatext "false" @default.
- W4382466690 isRetracted "false" @default.
- W4382466690 workType "article" @default.