Skip to main navigation Skip to search Skip to main content

Dynamic visual features based on discriminative speech class projection for visual speech recognition

  • Lei Xie
  • , Xiu Li Cai
  • , Zhong Hua Fu
  • , Rong Chun Zhao
  • Northwestern Polytechnical University Xian

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

This paper presents a dynamic visual feature extraction scheme to capture important lip motion information for visual speech recognition. Discriminative projections based on a-priori chosen speech classes, phonemes and visemes, are applied to the concatenation of pre-extracted static visual features. First- and second-order temporal derivatives are subsequently extracted to further represent the dynamic differences. Experiments on a connected digits task demonstrate that the proposed high discriminative dynamic features, when augmented to the static, yields superior recognition performances. Compared to the commonly used delta and acceleration features, the proposed dynamic feature leads to an 8% absolute improvement in terms of word accuracy for the considered recognition task.

Original languageEnglish
Title of host publication2004 International Symposium on Intelligent Multimedia, Video and Speech Processing, ISIMP 2004
Pages687-690
Number of pages4
StatePublished - 2004
Event2004 International Symposium on Intelligent Multimedia, Video and Speech Processing, ISIMP 2004 - Hong Kong, China, Hong Kong
Duration: 20 Oct 2004 → 22 Oct 2004

Publication series

Name2004 International Symposium on Intelligent Multimedia, Video and Speech Processing, ISIMP 2004

Conference

Conference2004 International Symposium on Intelligent Multimedia, Video and Speech Processing, ISIMP 2004
Country/TerritoryHong Kong
CityHong Kong, China
Period20/10/04 → 22/10/04

Fingerprint

Dive into the research topics of 'Dynamic visual features based on discriminative speech class projection for visual speech recognition'. Together they form a unique fingerprint.

Cite this