Face feature extraction method based on face feature point shape drive depth model

Abstract

The invention relates to a face feature extraction method based on a face feature point shape drive depth model. The method comprises the steps that the face feature point shape drive depth model is set up, N depth convolution neural networks are utilized for extracting features of N face regions divided according to the positions of face feature points to obtain the discrimination feature and the attributive feature of each region, and then all the discrimination features and the attributive features are fused to obtain features higher in descriptive ability. According to the face feature extraction method based on the face feature point shape drive depth model, the problem of robustness under change conditions of illumination, angles, expressions, shielding and the like can be well solved, and the recognition rate of face recognition under these conditions is increased.

Description

technical field [0001] The invention relates to a method for extracting human face features, in particular to a method for extracting human face features based on a shape-driven depth model of human face feature points. Background technique [0002] Face feature extraction is one of the most critical steps in face recognition. Before recognizing the target face image, it is first necessary to extract the face features in the image from multiple sample images of each face sample. The quality of face feature extraction will directly determine the effect of face recognition. Most of the existing face feature extraction methods are artificial feature extraction, such as SIFT, Gabor, HoG, LBP, etc. Further, better performance can be obtained by combining these artificial features. In recent years, feature extraction using deep learning has gradually become a research hotspot. Compared with artificial feature extraction methods, deep learning can obtain more effective features t...

AI Technical Summary

Problems solved by technology

Although the feature representation based on deep learning has brought a huge breakthrough for face recognition, face recognition under unconstrained conditions (changes in lighting, angle, expression, etc.) still has great challenges.

Moreover, the existing deep learning feature extraction methods usually only extract global features from the entire face image. Under the condition of local changes (changes in illumination, angle, expression, etc. will bring local changes), especially in the case of occlusion, recognition is prone to occur. error

Images