PROJECT TITLE :

Head Pose Estimation From a 2D Face Image Using 3D Face Morphing With Depth Parameters - 2015

ABSTRACT:

This paper presents estimation of head cause angles from one 2D face image using a 3D face model morphed from a reference face model. A reference model refers to a 3D face of a person of the identical ethnicity and gender as the question subject. The proposed theme minimizes the disparity between the 2 sets of distinguished facial options on the query face image and therefore the corresponding points on the 3D face model to estimate the pinnacle pose angles. The 3D face model used is morphed from a reference model to be a lot of specific to the query face in terms of the depth error at the feature points. The morphing method produces a 3D face model more specific to the query image when multiple 2D face pictures of the question subject are obtainable for training. The proposed morphing process is computationally economical since the depth of a 3D face model is adjusted by a scalar depth parameter at feature points. Optimal depth parameters are found by minimizing the disparity between the 2D options of the question face image and therefore the corresponding features on the morphed 3D model projected onto 2D space. The proposed head pose estimation technique was evaluated on two benchmarking databases: 1) the USF Human-ID database for depth estimation and a couple of) the Pointing'04 database for head cause estimation. Experiment results demonstrate that head create estimation errors in nodding and shaking angles are as low as seven.ninety three° and 4.sixty five° on average for one 2D input face image.