Research Papers

Mesh Processing for Computerized Facial Anthropometry

[+] Author and Article Information
Dhanannjay Deo

Center for Product Design and Manufacturing, Indian Institute of Science, Bangalore 560 012, Indiadeod@rpi.edu

Dibakar Sen

Center for Product Design and Manufacturing, Indian Institute of Science, Bangalore 560 012, Indiadibakar@cpdm.iisc.ernet.in



J. Comput. Inf. Sci. Eng 10(1), 011007 (Mar 10, 2010) (12 pages) doi:10.1115/1.3330420 History: Received February 15, 2008; Revised October 12, 2009; Published March 10, 2010; Online March 10, 2010

Understanding of the shape and size of different features of the human body from scanned data is necessary for automated design and evaluation of product ergonomics. In this paper, a computational framework is presented for automatic detection and recognition of important facial feature regions, from scanned head and shoulder polyhedral models. A noise tolerant methodology is proposed using discrete curvature computations, band-pass filtering, and morphological operations for isolation of the primary feature regions of the face, namely, the eyes, nose, and mouth. Spatial disposition of the critical points of these isolated feature regions is analyzed for the recognition of these critical points as the standard landmarks associated with the primary facial features. A number of clinically identified landmarks lie on the facial midline. An efficient algorithm for detection and processing of the midline, using a point sampling technique, is also presented. The results obtained using data of more than 20 subjects are verified through visualization and physical measurements. A color based and triangle skewness based schemes for isolation of geometrically nonprominent features and ear region are also presented.

Copyright © 2010 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Figure 1

Mesh preprocessing for filling holes: (a) raw mesh and (b) rectified mesh

Grahic Jump Location
Figure 2

Curvature of discrete surface: (a) circumcentric (Voronoi) regions, (b) barycentric (centroidal) regions, (c) parameters associated with edge xixj, and (d) area associated with a vertex in the mesh

Grahic Jump Location
Figure 3

Curvature distribution: Red for +ve curvature, blue for −ve curvature, and a darker shade for higher curvature; (a) Gaussian curvature and (b) mean curvature

Grahic Jump Location
Figure 4

Illustration of dilation, erosion, and opening operations on an object: (a) marked vertices (orange), (b) pink vertices added by dilation, (c) green vertex surviving erosion of (a), and (d) green vertices from erosion of (b) or opening of (a)

Grahic Jump Location
Figure 5

Steps to opening operation on face data: (a) mean curvature, (b) threshold [100,1000], (c) dilation, and (d) erosion

Grahic Jump Location
Figure 6

Mean curvature magnitude distribution of for a face: (a) whole face model and (b) frontal face data

Grahic Jump Location
Figure 7

Effect of lower threshold for morphological filtering on the selected regions

Grahic Jump Location
Figure 8

Effect of lower threshold on the number of feature regions on face

Grahic Jump Location
Figure 9

Identification and labeling of feature regions and landmarks

Grahic Jump Location
Figure 10

Facial feature regions and their extremities

Grahic Jump Location
Figure 11

Facial feature regions, extremities and centroids

Grahic Jump Location
Figure 12

Midline landmarks adopted from Ref. 50. Bold face annotations are maxima and italicized annotations are minima on the profile silhouette.

Grahic Jump Location
Figure 13

Different facial midline profiles and detected extremal points

Grahic Jump Location
Figure 14

Comparison of Y coordinate measure and distance from a reference point measure to detect extremal points on facial midline

Grahic Jump Location
Figure 15

Detected facial and midline features and landmarks on subject faces

Grahic Jump Location
Figure 16

Intensity distribution for fair and dark faces

Grahic Jump Location
Figure 17

Comparison of detected features and isolated moustache using intensity information for a typical face: (a) face as scanned, (b) data without color, (c) curvature based features, and (d) color based features

Grahic Jump Location
Figure 18

Filtering with triangle skewness. Skewness threshold is 2.5. (a) Close-up of triangles behind the ear, (b) frontal view, and (c) backside of ear.




Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In