Ghent University Academic Bibliography

Advanced

Segmented face approximation with adaptive region growing based on low-degree polynomial fitting

Francis Deboeverie, Peter Veelaert UGent and Wilfried Philips UGent (2015) SIGNAL IMAGE AND VIDEO PROCESSING. 9(2). p.347-363
abstract
Nowadays, an objective in visual communication is to send and store images of faces at a low bit rate, such that the faces are still recognizable and that the compression does not prevent remote face analysis. We present a novel segmented face approximation algorithm. Greyscale face images are segmented into meaningful surface segments with an adaptive region growing algorithm based on low-degree polynomial fitting. The algorithm uses a new adaptive thresholding technique with the L-infinity fitting cost as a segmentation criterion. The polynomial degree and the fitting error are automatically adapted during the region growing process. The main novelty is that the algorithm detects outliers and edges, distinguishes between strong and smooth intensity transitions and finds surface segments that are bent in a certain way, such as flat, planar, convex, concave or saddle patches. As a result, the surface segments correspond to facial features and the contours separating the surface segments coincide with real image face edges. Moreover, the curvature-based surface shape information facilitates many tasks in automated face analysis, demonstrated in this paper by face verification performed on the polynomial representation. The polynomial representation provides good approximation of facial features, while preserving all the necessary details of the face in the reconstructed image. When compared with different compression methods, we achieve higher compression ratios and better recognizable faces at low bit rates. This is confirmed by correct identification percentages obtained by face recognition algorithms on the compressed data.
Please use this url to cite or link to this publication:
author
organization
year
type
journalArticle (original)
publication status
published
subject
keyword
Region growing, Face recognition, Image approximation, Polynomial fitting, Face detection, Image segmentation, FEATURES, CONTEXT, RETRIEVAL, SPACE, VIDEO, ALGORITHM, K-SVD, COMPRESSION, FACIAL IMAGES, IMAGE-CODING TECHNIQUES
journal title
SIGNAL IMAGE AND VIDEO PROCESSING
SIViP
editor
Murat Kunt
volume
9
issue
2
pages
347 - 363
Web of Science type
Article
Web of Science id
000348112400008
JCR category
ENGINEERING, ELECTRICAL & ELECTRONIC
JCR impact factor
0.872 (2015)
JCR rank
167/255 (2015)
JCR quartile
3 (2015)
ISSN
1863-1703
DOI
10.1007/s11760-013-0441-6
language
English
UGent publication?
yes
classification
A1
copyright statement
I have transferred the copyright for this publication to the publisher
id
3143457
handle
http://hdl.handle.net/1854/LU-3143457
date created
2013-02-26 10:47:47
date last changed
2016-12-19 15:39:32
@article{3143457,
  abstract     = {Nowadays, an objective in visual communication is to send and store images of faces at a low bit rate, such that the faces are still recognizable and that the compression does not prevent remote face analysis. We present a novel segmented face approximation algorithm. Greyscale face images are segmented into meaningful surface segments with an adaptive region growing algorithm based on low-degree polynomial fitting. The algorithm uses a new adaptive thresholding technique with the L-infinity fitting cost as a segmentation criterion. The polynomial degree and the fitting error are automatically adapted during the region growing process. The main novelty is that the algorithm detects outliers and edges, distinguishes between strong and smooth intensity transitions and finds surface segments that are bent in a certain way, such as flat, planar, convex, concave or saddle patches. As a result, the surface segments correspond to facial features and the contours separating the surface segments coincide with real image face edges. Moreover, the curvature-based surface shape information facilitates many tasks in automated face analysis, demonstrated in this paper by face verification performed on the polynomial representation. The polynomial representation provides good approximation of facial features, while preserving all the necessary details of the face in the reconstructed image. When compared with different compression methods, we achieve higher compression ratios and better recognizable faces at low bit rates. This is confirmed by correct identification percentages obtained by face recognition algorithms on the compressed data.},
  author       = {Deboeverie, Francis and Veelaert, Peter and Philips, Wilfried},
  editor       = {Kunt, Murat},
  issn         = {1863-1703},
  journal      = {SIGNAL IMAGE AND VIDEO PROCESSING},
  keyword      = {Region growing,Face recognition,Image approximation,Polynomial fitting,Face detection,Image segmentation,FEATURES,CONTEXT,RETRIEVAL,SPACE,VIDEO,ALGORITHM,K-SVD,COMPRESSION,FACIAL IMAGES,IMAGE-CODING TECHNIQUES},
  language     = {eng},
  number       = {2},
  pages        = {347--363},
  title        = {Segmented face approximation with adaptive region growing based on low-degree polynomial fitting},
  url          = {http://dx.doi.org/10.1007/s11760-013-0441-6},
  volume       = {9},
  year         = {2015},
}

Chicago
Deboeverie, Francis, Peter Veelaert, and Wilfried Philips. 2015. “Segmented Face Approximation with Adaptive Region Growing Based on Low-degree Polynomial Fitting.” Ed. Murat Kunt. Signal Image and Video Processing 9 (2): 347–363.
APA
Deboeverie, F., Veelaert, P., & Philips, W. (2015). Segmented face approximation with adaptive region growing based on low-degree polynomial fitting. (M. Kunt, Ed.)SIGNAL IMAGE AND VIDEO PROCESSING, 9(2), 347–363.
Vancouver
1.
Deboeverie F, Veelaert P, Philips W. Segmented face approximation with adaptive region growing based on low-degree polynomial fitting. Kunt M, editor. SIGNAL IMAGE AND VIDEO PROCESSING. 2015;9(2):347–63.
MLA
Deboeverie, Francis, Peter Veelaert, and Wilfried Philips. “Segmented Face Approximation with Adaptive Region Growing Based on Low-degree Polynomial Fitting.” Ed. Murat Kunt. SIGNAL IMAGE AND VIDEO PROCESSING 9.2 (2015): 347–363. Print.