@inproceedings{afba2015d31c4a83aca558d3daa5895e,
title = "Generative cooperative net for image generation and data augmentation",
abstract = "How to build a good model for image generation given an abstract concept is one of fundamental problems in computer vision. In this paper, we explore a generative model for the task of generating fictitious images with desired features. We propose the Generative Cooperative Net (GCN) for image generation. The idea is similar to generative adversarial networks (GANs) except that the generators and discriminators are trained to work accordingly. We conducted experiments on hand-written digit generation and facial expression generation. In experimental studies, we found that GCN{\textquoteright}s two cooperative counterparts (the generator and the classifier) can work together nicely and achieve promising results. Such generative model can be used as a data-augmentation tool. Our experiment of applying this method to an emotion classification task shows that the synthesis images can help to improve classification accuracy. It is easy to set up and generate a very large synthesized dataset.",
keywords = "Data augmentation, GAN, Generative cooperative net (GCN)",
author = "Qiangeng Xu and Zengchang Qin and Tao Wan",
note = "Publisher Copyright: {\textcopyright} Springer Nature Switzerland AG 2019.; 7th International Symposium on Integrated Uncertainty in Knowledge Modelling and Decision Making, IUKM 2019 ; Conference date: 27-03-2019 Through 29-03-2019",
year = "2019",
doi = "10.1007/978-3-030-14815-7\_24",
language = "英语",
isbn = "9783030148140",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "284--294",
editor = "Van-Nam Huynh and Nguyen, \{Canh Hao\} and Hirosato Seki and Masahiro Inuiguchi",
booktitle = "Integrated Uncertainty in Knowledge Modelling and Decision Making - 7th International Symposium, IUKM 2019, Proceedings",
address = "德国",
}