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

Jie Qin, Yunhong Wang, Li Liu, Jiaxin Chen, Ling Shao

Research output: Contribution to journalArticlepeer-review

26 Citations (Scopus)


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.
Original languageEnglish
Pages (from-to)1667-1671
JournalIEEE Signal Processing Letters
Issue number11
Publication statusPublished - 21 Sept 2016


Dive into the research topics of 'Beyond Semantic Attributes: Discrete Latent Attributes Learning for Zero-Shot Recognition'. Together they form a unique fingerprint.

Cite this