Improving Semantic Embedding Consistency by Metric Learning for Zero-Shot Classification
Maxime Bucher  1@  , Frédéric Jurie  2, *@  , Stéphane Herbin  1, *@  
1 : Onera - The French Aerospace Lab  (Palaiseau)  -  Site web
ONERA
F-91761 Palaiseau -  France
2 : Groupe de Recherche en Informatique, Image, Automatique et Instrumentation de Caen  (GREYC)  -  Site web
Ecole Nationale Supérieure d'Ingénieurs de Caen, Université de Caen Basse-Normandie, CNRS : UMR6072
Boulevard du Maréchal Juin - 14050 CAEN Cedex -  France
* : Auteur correspondant

This paper addresses the task of zero-shot image classification. The key contribution of the proposed approach is to control the semantic embedding of images – one of the main ingredients of zero-shot learning – by formulating it as a metric learning problem. The optimized empirical criterion associates two types of sub-task constraints: metric discriminating capacity and accurate attribute prediction. This results in a novel expression of zero-shot learning not requiring the notion of class in the training phase: only pairs of image/attributes, augmented with a consistency indicator, are given as ground truth. At test time, the learned model can predict the consistency of a test image with a given set of attributes , allowing flexible ways to produce recognition inferences. Despite its simplicity, the proposed approach gives state-of-the-art results on four challenging datasets used for zero-shot recognition evaluation.


Personnes connectées : 1 Flux RSS