Generalized Zero-Shot Learning via Disentangled Representation

Authors

  • Xiangyu Li Xidian University
  • Zhe Xu Xidian University
  • Kun Wei Xidian University
  • Cheng Deng Xidian University

Keywords:

Language and Vision

Abstract

Zero-Shot Learning (ZSL) aims to recognize images belonging to unseen classes that are unavailable in the training process, while Generalized Zero-Shot Learning (GZSL) is a more realistic variant that both seen and unseen classes appear during testing. Most GZSL approaches achieve knowledge transfer based on the features of samples that inevitably contain information irrelevant to recognition, bringing negative influence for the performance. In this work, we propose a novel method, dubbed Disentangled-VAE, which aims to disentangle category-distilling factors and category-dispersing factors from visual as well as semantic features, respectively. In addition, a batch re-combining strategy on latent features is introduced to guide the disentanglement, encouraging the distilling latent features to be more discriminative for recognition. Extensive experiments demonstrate that our method outperforms the state-of-the-art approaches on four challenging benchmark datasets

Downloads

Published

2021-05-18

How to Cite

Li, X., Xu, Z., Wei, K., & Deng, C. (2021). Generalized Zero-Shot Learning via Disentangled Representation. Proceedings of the AAAI Conference on Artificial Intelligence, 35(3), 1966-1974. Retrieved from https://ojs.aaai.org/index.php/AAAI/article/view/16292

Issue

Section

AAAI Technical Track on Computer Vision II