Back to search

Clemson University

Visual Speech Recognition using Histogram of Oriented Displacements

Abstract

dc:description.abstract

Lip reading is the recognition of spoken words from the visual information of lips. It has been of considerable interest in the Computer Vision and Speech Recognition communities to automate this process using computer algorithms. In this thesis, we have developed a novel method involving describing visual features using fixed length descriptors called Histogram of Oriented Displacements to which we apply Support Vector Machines for recognition of spoken words. Using this method on the CUAVE database we have achieved a recognition rate of 81%.

Degree

thesis:*
Name thesis:degree_name
Master of Science (MS)
Level thesis:degree_level
Thesis
Discipline thesis:degree_discipline
Electrical Engineering

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Kumaravel, Sujeeth Selvam
Contributors dc:contributor
  • Dr. John Gowdy
  • Dr. Robert Schalkoff
  • Dr. Carl Baum

Subjects

dc:subject × 6

Identifiers

dc:identifier.*
Repository record dc:identifier
https://open.clemson.edu/all_theses/2112

Chain of custody

source
Harvested from
Clemson University
Base URL
open.clemson.edu/do/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Kumaravel, Sujeeth Selvam. Visual Speech Recognition using Histogram of Oriented Displacements. Thesis thesis, https://open.clemson.edu/all_theses/2112