Matches in SemOpenAlex for { <https://semopenalex.org/work/W2973286919> ?p ?o ?g. }
- W2973286919 abstract "Lately, generative adversarial networks (GANs) have been successfully applied to zero-shot learning (ZSL) and achieved state-of-the-art performance. By synthesizing virtual unseen visual features, GAN-based methods convert the challenging ZSL task into a supervised learning problem. However, GAN-based ZSL methods have to train the generator on the seen categories and further apply it to unseen instances. An inevitable issue of such a paradigm is that the synthesized unseen features are prone to seen references and incapable to reflect the novelty and diversity of real unseen instances. In a nutshell, the synthesized features are confusing. One cannot tell unseen categories from seen ones using the synthesized features. As a result, the synthesized features are too subtle to be classified in generalized zero-shot learning (GZSL) which involves both seen and unseen categories at the test stage. In this paper, we first introduce the feature confusion issue. Then, we propose a new feature generating network, named alleviating feature confusion GAN (AFC-GAN), to challenge the issue. Specifically, we present a boundary loss which maximizes the decision boundary of seen categories and unseen ones. Furthermore, a novel metric named feature confusion score (FCS) is proposed to quantify the feature confusion. Extensive experiments on five widely used datasets verify that our method is able to outperform previous state-of-the-arts under both ZSL and GZSL protocols." @default.
- W2973286919 created "2019-09-26" @default.
- W2973286919 creator A5008391423 @default.
- W2973286919 creator A5049692788 @default.
- W2973286919 creator A5057820981 @default.
- W2973286919 creator A5071936558 @default.
- W2973286919 creator A5078170935 @default.
- W2973286919 creator A5079281855 @default.
- W2973286919 date "2019-09-17" @default.
- W2973286919 modified "2023-09-24" @default.
- W2973286919 title "Alleviating Feature Confusion for Generative Zero-shot Learning" @default.
- W2973286919 cites W1797268635 @default.
- W2973286919 cites W2070148066 @default.
- W2973286919 cites W2099471712 @default.
- W2973286919 cites W2123024445 @default.
- W2973286919 cites W2124033848 @default.
- W2973286919 cites W2134270519 @default.
- W2973286919 cites W2171061940 @default.
- W2973286919 cites W2187089797 @default.
- W2973286919 cites W2194775991 @default.
- W2973286919 cites W2289084343 @default.
- W2973286919 cites W2398118205 @default.
- W2973286919 cites W2400717490 @default.
- W2973286919 cites W2405223529 @default.
- W2973286919 cites W2408719493 @default.
- W2973286919 cites W2435696002 @default.
- W2973286919 cites W2533598788 @default.
- W2973286919 cites W2552383788 @default.
- W2973286919 cites W2736590139 @default.
- W2973286919 cites W2739748921 @default.
- W2973286919 cites W2799215068 @default.
- W2973286919 cites W2887567284 @default.
- W2973286919 cites W2889143422 @default.
- W2973286919 cites W2892946488 @default.
- W2973286919 cites W2905461678 @default.
- W2973286919 cites W2927084868 @default.
- W2973286919 cites W2949503252 @default.
- W2973286919 cites W2950276680 @default.
- W2973286919 cites W2951313160 @default.
- W2973286919 cites W2951538594 @default.
- W2973286919 cites W2952142541 @default.
- W2973286919 cites W2955547856 @default.
- W2973286919 cites W2962879692 @default.
- W2973286919 cites W2963545832 @default.
- W2973286919 cites W2963960318 @default.
- W2973286919 cites W2964086552 @default.
- W2973286919 cites W652269744 @default.
- W2973286919 cites W93016980 @default.
- W2973286919 doi "https://doi.org/10.48550/arxiv.1909.07615" @default.
- W2973286919 hasPublicationYear "2019" @default.
- W2973286919 type Work @default.
- W2973286919 sameAs 2973286919 @default.
- W2973286919 citedByCount "2" @default.
- W2973286919 countsByYear W29732869192020 @default.
- W2973286919 countsByYear W29732869192021 @default.
- W2973286919 crossrefType "posted-content" @default.
- W2973286919 hasAuthorship W2973286919A5008391423 @default.
- W2973286919 hasAuthorship W2973286919A5049692788 @default.
- W2973286919 hasAuthorship W2973286919A5057820981 @default.
- W2973286919 hasAuthorship W2973286919A5071936558 @default.
- W2973286919 hasAuthorship W2973286919A5078170935 @default.
- W2973286919 hasAuthorship W2973286919A5079281855 @default.
- W2973286919 hasBestOaLocation W29732869191 @default.
- W2973286919 hasConcept C11171543 @default.
- W2973286919 hasConcept C119857082 @default.
- W2973286919 hasConcept C121332964 @default.
- W2973286919 hasConcept C12267149 @default.
- W2973286919 hasConcept C127413603 @default.
- W2973286919 hasConcept C138885662 @default.
- W2973286919 hasConcept C153180895 @default.
- W2973286919 hasConcept C154945302 @default.
- W2973286919 hasConcept C15744967 @default.
- W2973286919 hasConcept C163258240 @default.
- W2973286919 hasConcept C176217482 @default.
- W2973286919 hasConcept C21547014 @default.
- W2973286919 hasConcept C27206212 @default.
- W2973286919 hasConcept C2776401178 @default.
- W2973286919 hasConcept C2778738651 @default.
- W2973286919 hasConcept C2780992000 @default.
- W2973286919 hasConcept C2781140086 @default.
- W2973286919 hasConcept C39890363 @default.
- W2973286919 hasConcept C41008148 @default.
- W2973286919 hasConcept C41895202 @default.
- W2973286919 hasConcept C42023084 @default.
- W2973286919 hasConcept C62520636 @default.
- W2973286919 hasConceptScore W2973286919C11171543 @default.
- W2973286919 hasConceptScore W2973286919C119857082 @default.
- W2973286919 hasConceptScore W2973286919C121332964 @default.
- W2973286919 hasConceptScore W2973286919C12267149 @default.
- W2973286919 hasConceptScore W2973286919C127413603 @default.
- W2973286919 hasConceptScore W2973286919C138885662 @default.
- W2973286919 hasConceptScore W2973286919C153180895 @default.
- W2973286919 hasConceptScore W2973286919C154945302 @default.
- W2973286919 hasConceptScore W2973286919C15744967 @default.
- W2973286919 hasConceptScore W2973286919C163258240 @default.
- W2973286919 hasConceptScore W2973286919C176217482 @default.
- W2973286919 hasConceptScore W2973286919C21547014 @default.
- W2973286919 hasConceptScore W2973286919C27206212 @default.
- W2973286919 hasConceptScore W2973286919C2776401178 @default.
- W2973286919 hasConceptScore W2973286919C2778738651 @default.