Transductive Zero-Shot Learning with Visual Structure Constraint
Rattachement africain : gb, cn. Niveau de preuve : code pays fourni par la source.
Le résumé fourni par la source
To recognize objects of the unseen classes, most existing Zero-Shot Learning(ZSL) methods first learn a compatible projection function between the common semantic space and the visual space based on the data of source seen classes, then directly apply it to the target unseen classes. However, in real scenarios, the data distribution between the source and target domain might not match well, thus causing the well-known \textbf{domain shift} problem. Based on the observation that visual features of test instances can be separated into different clusters, we propose a new visual structure constraint on class centers for transductive ZSL, to improve the generality of the projection function (i.e. alleviate the above domain shift problem). Specifically, three different strategies (symmetric Chamfer-distance, Bipartite matching distance, and Wasserstein distance) are adopted to align the projected unseen semantic centers and visual cluster centers of test instances. We also propose a new training strategy to handle the real cases where many unrelated images exist in the test dataset, which is not considered in previous methods. Experiments on many widely used datasets demonstrate that the proposed visual structure constraint can bring substantial performance gain consistently and achieve state-of-the-art results. The source code is available at \url{https://github.com/raywzy/VSC}.
Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.
Le contrôle bibliographique ouvert
Où se fait cette recherche
-
Microsoft Research (United Kingdom) pays non établi dans la noticeEntreprise
-
Shenzhen University pays non établi dans la noticeUniversité ou école supérieure
-
Chinese Academy of Sciences pays non établi dans la noticeOrganisme public
-
(Microsoft) pays non établi dans la noticeInstitution
-
TENCENT pays non établi dans la noticeInstitution
-
Deepwise pays non établi dans la noticeInstitution
Microsoft Research (United Kingdom), Shenzhen University et Chinese Academy of Sciences, avec 3 autres affiliations.
Une affiliation ne permet pas de déduire la nationalité d’un auteur.