Zero-shot learning via discriminative representation extraction

Long, Teng, Xu, Xing, Shen, Fumin, Liu, Li, Xie, Ning and Yang, Yang (2018) Zero-shot learning via discriminative representation extraction. Pattern Recognition Letters, 109. pp. 27-34. ISSN 0167-8655

[thumbnail of Accepted manuscript]
PDF (Accepted manuscript) - Accepted Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (6MB) | Preview


Zero-shot learning (ZSL) aims to recognize classes whose samples did not appear during training. Existing research focuses on mapping deep visual feature to semantic embedding space explicitly or implicitly. However, ZSL improvements led by discriminative feature transformation is not well studied. In this paper, we propose a ZSL framework that maps semantic embeddings to a discriminative representation space, which are learned in two different ways: Kernelized Linear Discriminant Analysis (KLDA) and Central-loss based Network (CLN). KLDA and CLN can both force samples to be intra-class aggregation and inter-class separation. With the learned discriminative representations, we map class embeddings to representation space using Kernelized Ridge Regression (KRR). Our experiments show that both KLDA+KRR and CLN+KRR surpass state-of-art approaches in both recognition and retrieval task.

Item Type: Article
Uncontrolled Keywords: zero shot learning,large margin,aggregation,representation learning
Faculty \ School: Faculty of Science > School of Computing Sciences
Related URLs:
Depositing User: Pure Connector
Date Deposited: 29 Sep 2017 05:06
Last Modified: 22 Oct 2022 03:13
DOI: 10.1016/j.patrec.2017.09.030

Actions (login required)

View Item View Item