Neighborhood-Preserving estimation algorithm for facial landmark points

Authors:
Yin Liu;Ying Cui;Zhong Jin
Affiliations:
School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China;School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China;School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China
Venue:
IScIDE'12 Proceedings of the third Sino-foreign-interchange conference on Intelligent Science and Intelligent Data Engineering
Year:
2012

Citing 6
Cited 0

Images as Bags of Pixels

ICCV '03 Proceedings of the Ninth IEEE International Conference on Computer Vision - Volume 2
Automatic Construction of Active Appearance Models as an Image Coding Problem

IEEE Transactions on Pattern Analysis and Machine Intelligence
A 3D Facial Expression Database For Facial Behavior Research

FGR '06 Proceedings of the 7th International Conference on Automatic Face and Gesture Recognition
Learning active appearance models from image sequences

VisHCI '06 Proceedings of the HCSNet workshop on Use of vision in human-computer interaction - Volume 56
Automatic frontal face annotation and AAM building for arbitrary expressions from a single frontal image only

ICIP'09 Proceedings of the 16th IEEE international conference on Image processing
Regression based automatic face annotation for deformable model building

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we propose a neighborhood-preserving estimation (NPE) algorithm for facial landmark points at arbitrary poses. The proposed NPE algorithm is based on the following assumption: the neighboring structure of the face shapes in non-frontal view is consistent with that in frontal view. A face shape both in frontal and non-frontal view is represented as a linear combination of its neighbors. It is assumed the neighbors both in frontal and non-frontal view are from same persons and with same weights in the combinations. Extensive experiments are conducted on IMM and BU_3DFE to validate the performance of the proposed algorithm.