Beyond Semantic Attributes: Discrete Latent Attributes Learning for Zero-Shot Recognition

Qin, Jie, Wang, Yunhong, Liu, Li, Chen, Jiaxin and Shao, Ling (2016) Beyond Semantic Attributes: Discrete Latent Attributes Learning for Zero-Shot Recognition. IEEE Signal Processing Letters, 23 (11). pp. 1667-1671. ISSN 1070-9908

Full text not available from this repository. (Request a copy)
Official URL:


In this letter, we propose a novel approach for learning semantics-driven attributes, which are discriminative for zero-shot visual recognition. Latent attributes are derived in a principled manner, aiming at maintaining class-level semantic relatedness and attribute-wise balancedness. Unlike existing methods that binarize learned real-valued attributes via a quantization stage, we directly learn the optimal binary attributes by effectively addressing a discrete optimization problem. Particularly, we propose a class-wise discrete descent algorithm, based on which latent attributes of each class are learned iteratively. Moreover, we propose to simultaneously predict multiple attributes from low-level features via multioutput neural networks (MONN), which can model intrinsic correlation among attributes and make prediction more tractable. Extensive experiments on two standard datasets clearly demonstrate the superiority of our method over the state-of-the-arts.

Item Type: Article
Uncontrolled Keywords: Class-level attribute learning, discrete optimization, multioutput neural networks, zero-shot learning
Subjects: G400 Computer Science
Department: Faculties > Engineering and Environment > Computer and Information Sciences
Depositing User: Ellen Cole
Date Deposited: 07 Nov 2016 14:03
Last Modified: 12 Oct 2019 20:46

Actions (login required)

View Item View Item


Downloads per month over past year

View more statistics