Supervised local linear embedding (SLLE) for facial paralysis image sequence analysis

Research output: Chapter in Book/Report/Conference proceedingChapter

1 Citation (Scopus)

Abstract

This paper proposed a novel approach based on local linear embedding (LLE) for modeling and understanding the temporal behaviour from facial paralysis image sequences. LLE is sensitive to scaling, illumination and face pose. A Supervised LLE (SLLE) based on directed Hausdorff distance is proposed for aligning two image sets from the two sides of the face using one generalized embedding space. The embedded low dimensional manifolds represent the asymmetry of the facial motion regardless of the extrinsic facial asymmetry caused by the natural bilateral asymmetry, illumination and shadows. Experimental results demonstrate that our approach is more reliable than frame difference based methods and optical flow based methods without a significant increase in computational complexity.
LanguageEnglish
Title of host publication2008 IEEE International Conference on Multimedia and Expo
PublisherIEEE
Pages49-52
Number of pages3
ISBN (Print)978-1-4244-2570-9
DOIs
Publication statusPublished - Jun 2008

Fingerprint

Lighting
Optical flows
Computational complexity

Keywords

  • computational complexity
  • image sequences
  • medical image processing
  • Hausdorff distance
  • extrinsic facial asymmetry
  • facial paralysis image sequence analysis
  • frame difference based methods
  • generalized embedding space
  • natural bilateral asymmetry
  • optical flow based methods
  • supervised local linear embedding

Cite this

He, S., & Soraghan, J. J. (2008). Supervised local linear embedding (SLLE) for facial paralysis image sequence analysis. In 2008 IEEE International Conference on Multimedia and Expo (pp. 49-52). IEEE. https://doi.org/10.1109/ICME.2008.4607368
He, Shu ; Soraghan, J.J. / Supervised local linear embedding (SLLE) for facial paralysis image sequence analysis. 2008 IEEE International Conference on Multimedia and Expo. IEEE, 2008. pp. 49-52
@inbook{29e705bb024d4b97ac5bd08401b4bf50,
title = "Supervised local linear embedding (SLLE) for facial paralysis image sequence analysis",
abstract = "This paper proposed a novel approach based on local linear embedding (LLE) for modeling and understanding the temporal behaviour from facial paralysis image sequences. LLE is sensitive to scaling, illumination and face pose. A Supervised LLE (SLLE) based on directed Hausdorff distance is proposed for aligning two image sets from the two sides of the face using one generalized embedding space. The embedded low dimensional manifolds represent the asymmetry of the facial motion regardless of the extrinsic facial asymmetry caused by the natural bilateral asymmetry, illumination and shadows. Experimental results demonstrate that our approach is more reliable than frame difference based methods and optical flow based methods without a significant increase in computational complexity.",
keywords = "computational complexity, image sequences, medical image processing, Hausdorff distance, extrinsic facial asymmetry, facial paralysis image sequence analysis, frame difference based methods, generalized embedding space, natural bilateral asymmetry, optical flow based methods, supervised local linear embedding",
author = "Shu He and J.J. Soraghan",
year = "2008",
month = "6",
doi = "10.1109/ICME.2008.4607368",
language = "English",
isbn = "978-1-4244-2570-9",
pages = "49--52",
booktitle = "2008 IEEE International Conference on Multimedia and Expo",
publisher = "IEEE",

}

Supervised local linear embedding (SLLE) for facial paralysis image sequence analysis. / He, Shu; Soraghan, J.J.

2008 IEEE International Conference on Multimedia and Expo. IEEE, 2008. p. 49-52.

Research output: Chapter in Book/Report/Conference proceedingChapter

TY - CHAP

T1 - Supervised local linear embedding (SLLE) for facial paralysis image sequence analysis

AU - He, Shu

AU - Soraghan, J.J.

PY - 2008/6

Y1 - 2008/6

N2 - This paper proposed a novel approach based on local linear embedding (LLE) for modeling and understanding the temporal behaviour from facial paralysis image sequences. LLE is sensitive to scaling, illumination and face pose. A Supervised LLE (SLLE) based on directed Hausdorff distance is proposed for aligning two image sets from the two sides of the face using one generalized embedding space. The embedded low dimensional manifolds represent the asymmetry of the facial motion regardless of the extrinsic facial asymmetry caused by the natural bilateral asymmetry, illumination and shadows. Experimental results demonstrate that our approach is more reliable than frame difference based methods and optical flow based methods without a significant increase in computational complexity.

AB - This paper proposed a novel approach based on local linear embedding (LLE) for modeling and understanding the temporal behaviour from facial paralysis image sequences. LLE is sensitive to scaling, illumination and face pose. A Supervised LLE (SLLE) based on directed Hausdorff distance is proposed for aligning two image sets from the two sides of the face using one generalized embedding space. The embedded low dimensional manifolds represent the asymmetry of the facial motion regardless of the extrinsic facial asymmetry caused by the natural bilateral asymmetry, illumination and shadows. Experimental results demonstrate that our approach is more reliable than frame difference based methods and optical flow based methods without a significant increase in computational complexity.

KW - computational complexity

KW - image sequences

KW - medical image processing

KW - Hausdorff distance

KW - extrinsic facial asymmetry

KW - facial paralysis image sequence analysis

KW - frame difference based methods

KW - generalized embedding space

KW - natural bilateral asymmetry

KW - optical flow based methods

KW - supervised local linear embedding

UR - http://dx.doi.org/10.1109/ICME.2008.4607368

U2 - 10.1109/ICME.2008.4607368

DO - 10.1109/ICME.2008.4607368

M3 - Chapter

SN - 978-1-4244-2570-9

SP - 49

EP - 52

BT - 2008 IEEE International Conference on Multimedia and Expo

PB - IEEE

ER -