Three-dimensional face identification device and method based on three-dimensional point cloud

Abstract

The invention discloses a three-dimensional face identification device and method based on the three-dimensional point cloud. The device comprises a characteristic region detection unit for carrying out positioning on a characteristic region of the three-dimensional point cloud; a depth image mapping unit for carrying out normalization mapping on the three-dimensional point cloud to a depth image space; a Gabor response computation unit for carrying out different-dimension and different-direction response computation on three-dimensional face data by utilizing different-dimension and different-direction Gabor filters; a storage unit for storing visual dictionary of the three-dimensional face data obtained by training; and a histogram mapping computation unit for carrying out histogram mapping with the visual dictionary, for the Gabor response vector obtained by each pixel. To begin with, positioning and registeration are carried out on the characteristic region of a three-dimensional face region; then, point cloud data is mapped to a depth image according to the depth information; next, visual dictionary histogram vector of the three-dimensional data is carried out according to the trained three-dimensional face visual dictionary; and finally, identification can be realized through a classifier, and the identification precision is high.

Description

Technical field [0001] The present invention relates to the technical field of face recognition, in particular to a three-dimensional face recognition device and method based on a three-dimensional point cloud. Background technique [0002] Compared with two-dimensional face recognition, three-dimensional face recognition has the advantages of being robust to lighting and less affected by factors such as gestures and expressions. Therefore, after the rapid development of three-dimensional data collection technology and the greatly improved quality and accuracy of three-dimensional data Many researchers have focused their research into this field. [0003] Among them, the Chinese invention patent application No. 201010256907.6 published on December 29, 2010 proposes a method for describing the characteristics of the face by using the relevant features of three-dimensional bending invariants. This method encodes the curvature of the adjacent nodes on the three-dimensional face surfac...

AI Technical Summary

Problems solved by technology

[0005] 3D face recognition is the basic work of many applications in the field of 3D face. Most of the initial work in this field is to use the information of 3D data: such as curvature, depth and other data to describe the face, but due to the collection of 3D data There are a lot of noise points in the data, so the characteristic data such as curvature are sensitive to noise, so that the feature description vector of the three-dimensional face is not accurate in the recognition result.

After mapping the three-dimensional data to the depth map data, many appearance features of two-dimensional faces have been applied to this field, such as principal component analysis (PCA) and Gabor (Gabor) filter features; but these features also have their own shortcomings: (1) For the PCA feature, because it belongs to the global appearance feature, the ability to describe the detailed texture of the 3D data is insufficient; (2) For the Gabor filter feature, due to the noise problem of the 3D data, it is not suitable for the 3D face. The descriptive ability of the data depends on the quality of the acquired 3D face data

Images