A Face Alignment Method Based on Coarse-to-fine Face Shape Estimation

Abstract

The invention discloses a face alignment method based on face shape estimation from coarse to fine. For any input face picture, the initial face shape is first estimated, and then the real shape of the face is gradually approached, including using multiple The deep learning framework of the task estimates the position of the main feature points and facial expressions of the face, builds a head pose classification model based on convolutional neural network to accurately estimate and classify the head pose of the face, and uses the head pose classification results As well as the estimation results of facial expressions and the positions of main feature points, a more accurate initial shape can be obtained; based on the initial shape, according to the classification results of pose and expression, the respective regressors are trained to update the face shape to approach the standard shape. The present invention improves the robustness to differences in facial expressions, head poses, and light occlusion by constructing more accurate face initialization shapes and adopting a more advanced cascade regression framework.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and in particular relates to a face alignment method based on rough-to-fine face shape estimation in the field of face recognition of digital images. Background technique [0002] Face alignment can provide accurate and semantically specific face shape information, which can help realize geometric image normalization and feature extraction. Therefore, face alignment is an indispensable and important part of face recognition, facial posture and expression analysis, human-computer interaction and 3D face modeling. It is widely used in security, public security control, intelligent access control, human-computer interaction, assisted driving, Film and television production, video conferencing and other fields. In practice, the face alignment problem still faces great challenges due to differences in facial expressions, head poses, lighting conditions, and partial occlusions. Therefore, how...

AI Technical Summary

Problems solved by technology

[0031] In order to solve the above-mentioned technical problems, the present invention proposes a face alignment method based on face shape estimation from coarse to fine, which mainly solves the problems of facial expression, head posture, differences in lighting conditions and partial occlusion. Alignment accuracy is not high

Images