Recent work has shown the advantages of using attribute-based representation over low-level feature descriptors in face verification, due to its capability to explicitly encode high-level semantic meaning with economical coding bits. However, most work assumes that the attributes of a given face is independent to each other. In this paper we present a novel method to show how to model the relationship between attributes and exploit such information in the task of face verification, while taking uncertainty in attribute responses into account. Specifically, inspired by the vector representation of words in the literature of text categorization, we first represent the meaning of each attribute as a highdimensional vector in the subject space, then construct an attribute-relationship graph based on the distribution of attributes in that space. Using this, we are able to explicitly constrain the searching space of parameter values of a discriminative classifier to avoid over-fitting. The effectiveness of the proposed method is verified on the challenging Labeled Face in the Wild (LFW) database with promising results.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.