Matches in SemOpenAlex for { <https://semopenalex.org/work/W4381480456> ?p ?o ?g. }
- W4381480456 endingPage "126463" @default.
- W4381480456 startingPage "126463" @default.
- W4381480456 abstract "It is usually supposed that training database is manually balanced in traditional visual recognition tasks. However, in nature, data tends to follow long-tailed distributions. In recent years, many plug-and-play methods based on data augmentation or representation learning have been proposed to tackle the long-tailed visual recognition task. Although these methods are effective, we find that when different plug-and-play methods are applied to the same long-tail recognition model, they sometimes fail to promote each other. The reason for this phenomenon may lie in the fact that the overall performance of the model is constrained by the insufficient capability of a traditional feature extractor. Motivated by this fact, we first propose Hierarchical Block Aggregation Network (HBAN), a network structure with stronger feature extraction capability. Then, we design a Quantity-Aware Balanced (QAB) loss and a decoupled training paradigm to optimize HBAN. The effectiveness of HBAN is demonstrated by extensive experiments. In particular, HBAN achieves significant improvements over our baseline on three benchmark datasets, and outperforms the state-of-the-art methods on CIFAR 100-LT." @default.
- W4381480456 created "2023-06-22" @default.
- W4381480456 creator A5006557413 @default.
- W4381480456 creator A5016523761 @default.
- W4381480456 creator A5041227155 @default.
- W4381480456 creator A5081619260 @default.
- W4381480456 date "2023-09-01" @default.
- W4381480456 modified "2023-09-27" @default.
- W4381480456 title "Hierarchical block aggregation network for long-tailed visual recognition" @default.
- W4381480456 cites W1834627138 @default.
- W4381480456 cites W1861492603 @default.
- W4381480456 cites W2108598243 @default.
- W4381480456 cites W2119168155 @default.
- W4381480456 cites W2163922914 @default.
- W4381480456 cites W2194321275 @default.
- W4381480456 cites W2194775991 @default.
- W4381480456 cites W2549139847 @default.
- W4381480456 cites W2797977484 @default.
- W4381480456 cites W2962858109 @default.
- W4381480456 cites W2962933664 @default.
- W4381480456 cites W2992308087 @default.
- W4381480456 cites W3034711780 @default.
- W4381480456 cites W3035552357 @default.
- W4381480456 cites W3096688134 @default.
- W4381480456 cites W3176474016 @default.
- W4381480456 cites W3177230409 @default.
- W4381480456 cites W3178139507 @default.
- W4381480456 cites W3215720729 @default.
- W4381480456 cites W4214640945 @default.
- W4381480456 cites W4214693531 @default.
- W4381480456 cites W4214718285 @default.
- W4381480456 cites W4291221943 @default.
- W4381480456 cites W4309478457 @default.
- W4381480456 cites W4312244159 @default.
- W4381480456 cites W4312702685 @default.
- W4381480456 cites W4312739324 @default.
- W4381480456 cites W4312877479 @default.
- W4381480456 cites W4316039445 @default.
- W4381480456 doi "https://doi.org/10.1016/j.neucom.2023.126463" @default.
- W4381480456 hasPublicationYear "2023" @default.
- W4381480456 type Work @default.
- W4381480456 citedByCount "1" @default.
- W4381480456 countsByYear W43814804562023 @default.
- W4381480456 crossrefType "journal-article" @default.
- W4381480456 hasAuthorship W4381480456A5006557413 @default.
- W4381480456 hasAuthorship W4381480456A5016523761 @default.
- W4381480456 hasAuthorship W4381480456A5041227155 @default.
- W4381480456 hasAuthorship W4381480456A5081619260 @default.
- W4381480456 hasConcept C117978034 @default.
- W4381480456 hasConcept C119857082 @default.
- W4381480456 hasConcept C124101348 @default.
- W4381480456 hasConcept C127413603 @default.
- W4381480456 hasConcept C13280743 @default.
- W4381480456 hasConcept C138885662 @default.
- W4381480456 hasConcept C153180895 @default.
- W4381480456 hasConcept C154945302 @default.
- W4381480456 hasConcept C162324750 @default.
- W4381480456 hasConcept C17744445 @default.
- W4381480456 hasConcept C185798385 @default.
- W4381480456 hasConcept C187736073 @default.
- W4381480456 hasConcept C199539241 @default.
- W4381480456 hasConcept C205649164 @default.
- W4381480456 hasConcept C21880701 @default.
- W4381480456 hasConcept C2524010 @default.
- W4381480456 hasConcept C2776359362 @default.
- W4381480456 hasConcept C2776401178 @default.
- W4381480456 hasConcept C2777210771 @default.
- W4381480456 hasConcept C2780451532 @default.
- W4381480456 hasConcept C33923547 @default.
- W4381480456 hasConcept C41008148 @default.
- W4381480456 hasConcept C41895202 @default.
- W4381480456 hasConcept C52622490 @default.
- W4381480456 hasConcept C94625758 @default.
- W4381480456 hasConceptScore W4381480456C117978034 @default.
- W4381480456 hasConceptScore W4381480456C119857082 @default.
- W4381480456 hasConceptScore W4381480456C124101348 @default.
- W4381480456 hasConceptScore W4381480456C127413603 @default.
- W4381480456 hasConceptScore W4381480456C13280743 @default.
- W4381480456 hasConceptScore W4381480456C138885662 @default.
- W4381480456 hasConceptScore W4381480456C153180895 @default.
- W4381480456 hasConceptScore W4381480456C154945302 @default.
- W4381480456 hasConceptScore W4381480456C162324750 @default.
- W4381480456 hasConceptScore W4381480456C17744445 @default.
- W4381480456 hasConceptScore W4381480456C185798385 @default.
- W4381480456 hasConceptScore W4381480456C187736073 @default.
- W4381480456 hasConceptScore W4381480456C199539241 @default.
- W4381480456 hasConceptScore W4381480456C205649164 @default.
- W4381480456 hasConceptScore W4381480456C21880701 @default.
- W4381480456 hasConceptScore W4381480456C2524010 @default.
- W4381480456 hasConceptScore W4381480456C2776359362 @default.
- W4381480456 hasConceptScore W4381480456C2776401178 @default.
- W4381480456 hasConceptScore W4381480456C2777210771 @default.
- W4381480456 hasConceptScore W4381480456C2780451532 @default.
- W4381480456 hasConceptScore W4381480456C33923547 @default.
- W4381480456 hasConceptScore W4381480456C41008148 @default.
- W4381480456 hasConceptScore W4381480456C41895202 @default.
- W4381480456 hasConceptScore W4381480456C52622490 @default.
- W4381480456 hasConceptScore W4381480456C94625758 @default.