Multi-view two-dimension facial feature point automatic positioning method

Abstract

The invention relates to the technical field of the computer applications and the computer view, in particular to a multi-view two-dimension facial feature point automatic positioning method. The multi-view two-dimension facial feature point automatic positioning method comprises training and testing stages; the training stage comprises the following steps of firstly dividing a training data set containing a multi-view facial image into a plurality of training sub-sets; secondly training a single-view feature point positioning engine, namely training a cascading regression feature point positioning engine aiming at every training sub-set in every step. The multi-view two-dimension facial feature point automatic positioning method improves the robustness of feature point positioning under every view and can detect feature points of two-dimension facial images with the horizontal deflection angle from minus 90 degrees to 90 degrees.

Description

technical field [0001] The invention relates to the fields of computer application technology and computer vision technology, in particular to a multi-view two-dimensional human face feature point automatic positioning method. Background technique [0002] Face feature points (such as nose tip, pupil center, mouth corner, etc.) play a very important role in many problems related to the face. For example, in face recognition, face feature points are widely used in face alignment, scale normalization Synthesis and feature template extraction, the facial shape defined by facial feature points in facial expression analysis is an important basis for expression changes. Therefore, in the past ten years, face feature point location has attracted the attention of a large number of researchers, and various methods have been proposed. [0003] Existing face landmark localization methods can be broadly divided into two categories: methods based on statistical shape models and methods ...

AI Technical Summary

Problems solved by technology

However, these methods did not distinguish face images from different perspectives during the training process. As a result, although they work well on face images that are frontal and close to the front (within 45 degrees of left and right deflection), they do not work well on face images with a large posture deflection angle. Detection accuracy drops rapidly on face images

Images