INTERNATIONAL JOURNAL OF SCIENTIFIC DEVELOPMENT AND RESEARCH International Peer Reviewed & Refereed Journals, Open Access Journal ISSN Approved Journal No: 2455-2631 | Impact factor: 8.15 | ESTD Year: 2016
open access , Peer-reviewed, and Refereed Journals, Impact factor 8.15
Recognition of Dynamic Facial Expression Using Atlas Construction and Sparsity Orthogonal Matching
Authors Name:
Ashwini Sugave
, Ashwini Gaikwad
Unique Id:
IJSDR1812048
Published In:
Volume 3 Issue 11, December-2018
Abstract:
Human face can exhibits complex and strong changes that are both unpredictable and varying in time Now a days, facial expression recognition has become a emerging and research topic due to ad-vancement in this field. To overcome the disadvantages of conventional group-wise registration methods Spatio-temporal information and SOM algorithm used in recognition stage.This paper presents sparse representation with sparsity estimation and variable step size to improve recognition efficiency and accuracy. We have adopted SOM algorithm for better performance with reference to sparse representation classification algorithm. We also explored Sparsity Orthogonal matching algorithm is used in atlas construction to reach to true sparsity and lead to improved sparse representation as well as to resolve sparse coefficient and larger step size to shorten the time of iterations. The recognition rate is higher than other method shows by experimental result.
"Recognition of Dynamic Facial Expression Using Atlas Construction and Sparsity Orthogonal Matching", International Journal of Science & Engineering Development Research (www.ijsdr.org), ISSN:2455-2631, Vol.3, Issue 11, page no.285 - 290, December-2018, Available :http://www.ijsdr.org/papers/IJSDR1812048.pdf
Downloads:
000337350
Publication Details:
Published Paper ID: IJSDR1812048
Registration ID:180903
Published In: Volume 3 Issue 11, December-2018
DOI (Digital Object Identifier):
Page No: 285 - 290
Publisher: IJSDR | www.ijsdr.org
ISSN Number: 2455-2631
Facebook Twitter Instagram LinkedIn