Novel Technique for 3D Face Segmentation and Landmarking

In this paper we propose a new technique to detect and extract any part that we need from a 3D face. The regions of interest of our approach are the eyes and the nose. Moreover, we are interested in the extraction of the salient landmarks in 3D face. Therefore, in this paper a new method is proposed...

Full description

Saved in:
Bibliographic Details
Published in2016 Global Summit on Computer & Information Technology (GSCIT) pp. 27 - 31
Main Authors Sghaier, Souhir, Souani, Chokri, Faeidh, Hassene, Besbes, Kamel
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.07.2016
Subjects
3D face
anthropometric
Face
Image resolution
Information technology
landmarks
Noise measurement
Nose
region of interest
Robustness
segmentation
Three-dimensional displays
Online AccessGet full text

Cover

More Information
Summary:In this paper we propose a new technique to detect and extract any part that we need from a 3D face. The regions of interest of our approach are the eyes and the nose. Moreover, we are interested in the extraction of the salient landmarks in 3D face. Therefore, in this paper a new method is proposed to detect the nose tip and eye corners from a three-dimensional face range image. Our original technique allows fully automated processing, treating incomplete and noisy input data, and automatically rejecting non-facial areas. Besides, it is robust against holes in 3D image and insensitive to facial expressions. Besides, it is stable against translation and rotation of the face, and it is suitable for different resolutions of images. All the experiments have been performed on the GAVAB and FRAV 3D databases. After applying the proposed method to the 3D face, experimental results show that it is comparable to state-of-the-art methods in terms of its accuracy and flexibility.
DOI:10.1109/GSCIT.2016.17