Journal
IEEE TRANSACTIONS ON MULTIMEDIA
Volume 24, Issue -, Pages 1325-1337Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TMM.2021.3063616
Keywords
Generalized zero-shot learning; domain shift problem; generative adversarial networks; latent attributes; seen/unseen classifier
Categories
Funding
- National Natural Science Foundation of China [61976047, 61872061, 61806039]
- Science, and Technology Department of Sichuan Province of China [2019YFG0122, 2020YFG0087, 2020YFG0080, 2020YFG0326]
Ask authors/readers for more resources
This paper proposes a generative model called discriminative learning GAN (DL-GAN) to alleviate the domain shift problem in generalized zero-shot learning (GZSL). DL-GAN utilizes a dual-stream embedding model, an attribute-based generative model, and a seen/unseen classifier to improve performance. Experimental results demonstrate that our proposed approach significantly outperforms existing methods on four datasets under the GZSL protocol.
In zero-shot learning (ZSL) tasks, especially in generalized zero-shot learning (GZSL), the model tends to classify unseen test samples into seen categories, which is well known as the domain shift problem, because the model is trained from seen samples without unseen samples. Recently, generative adversarial network (GAN) based methods have achieved good performance in GZSL, which replace real unseen features by synthesizing fake ones to mitigate the domain shift. However, the domain shift problem is still not well solved, due to the lacking of unseen samples in the training progress of the GAN generator. In this paper, we propose a generative model named discriminative learning GAN (DL-GAN) to alleviate the domain shift in GZSL. Specifically, the DL-GAN is designed with three novel components: a dual-stream embedding model that aligns features to the ground-truth attributes to extract discriminative latent attributes from features, an attribute-based generative model that generates high-quality unseen features from semantic attributes to guarantee inter-class discriminability and semantic consistency, and a seen/unseen classifier that leverages validation samples to distinguish seen samples from unseen ones. Experimental results on four widely used datasets verify that our proposed approach significantly outperforms the state-of-the-art methods under the GZSL protocol.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available