Two features combination with gated recurrent unit for visual speech recognition

Abstract

Recently, deep learning has attracted in the visual speech recognition (VSR) field, however, most of approaches use single feature. This paper proposes two features; image-based feature by autoencoder, and motion-based feature based on feature points, and combination approach with gated recurrent unit for VSR task. The proposed method was evaluated on a… (More)
DOI: 10.23919/MVA.2017.7986867

Topics

9 Figures and Tables

Cite this paper

@article{Iwasaki2017TwoFC, title={Two features combination with gated recurrent unit for visual speech recognition}, author={Masaya Iwasaki and Michiko Kubokawa and Takeshi Saitoh}, journal={2017 Fifteenth IAPR International Conference on Machine Vision Applications (MVA)}, year={2017}, pages={326-329} }