English

Cross-modal Variational Auto-encoder with Distributed Latent Spaces and Associators

Machine Learning 2019-05-31 v1 Machine Learning

Abstract

In this paper, we propose a novel structure for a cross-modal data association, which is inspired by the recent research on the associative learning structure of the brain. We formulate the cross-modal association in Bayesian inference framework realized by a deep neural network with multiple variational auto-encoders and variational associators. The variational associators transfer the latent spaces between auto-encoders that represent different modalities. The proposed structure successfully associates even heterogeneous modal data and easily incorporates the additional modality to the entire network via the proposed cross-modal associator. Furthermore, the proposed structure can be trained with only a small amount of paired data since auto-encoders can be trained by unsupervised manner. Through experiments, the effectiveness of the proposed structure is validated on various datasets including visual and auditory data.

Keywords

Cite

@article{arxiv.1905.12867,
  title  = {Cross-modal Variational Auto-encoder with Distributed Latent Spaces and Associators},
  author = {Dae Ung Jo and ByeongJu Lee and Jongwon Choi and Haanju Yoo and Jin Young Choi},
  journal= {arXiv preprint arXiv:1905.12867},
  year   = {2019}
}

Comments

10 pages, 6 figures

R2 v1 2026-06-23T09:32:41.916Z