Method identifying non-front-side facial expression based on attitude normalization

Abstract

The invention discloses a method identifying a non-front-side facial expression based on attitude normalization. The method comprises that facial expressions in a training sample set are learned via a nonlinear regression model to obtain a mapping function from non-front-side facial characteristic points to front-side facial characteristic points; attitude estimation and characteristic point positioning are carried out on to-be-tested non-front-side facial images via a multi-template active appearance model, and the characteristic points of a non-front-side face are normalized to a front-side attitude via the corresponding attitude mapping function; and geometric positions of the characteristic points of a front-side face are classified into expressions via a support vector machine. The method identifying the non-front-side facial expression is simple and effectively, which solves the problem that different facial attitudes cause different expressions, and satisfies the requirement of identifying non-front-side facial expressions in real time.

Description

technical field [0001] The invention relates to the field of pattern recognition and image processing, in particular to a non-frontal facial expression recognition method based on gesture normalization. Background technique [0002] Expressions are the external manifestations of emotions and emotions. According to the basic emotion model, expressions can be divided into six categories: anger, disgust, fear, happiness, sadness, and surprise. Facial expression recognition has always been of great research significance, and has huge market value in many fields such as human-computer interaction, public safety, and intelligent film and television. Most of the traditional expression recognition methods take frontal or near-frontal face images as the research objects. However, statistical results show that due to the randomness of image acquisition in real life, 75% of face images are non-positive. If we only use traditional methods to analyze these non-frontal face images, we o...

AI Technical Summary

Problems solved by technology

[0003] Compared with the frontal facial expression images, most of the non-frontal facial expression images have part of the face covered, resulting in the lack of certain expression information; at the same time, the diversity of facial posture changes will inevitably introduce greater categories to the expression classification. In addition, it is very difficult to find expression features independent of facial gestures, and the features extracted by traditional frontal expression recognition methods will introduce a lot of redundant information including gesture changes

Images