Lightweight human face 3D key point detection method and system

Abstract

The invention discloses a lightweight human face 3D key point detection method and system. The method comprises the steps of performing dimension reduction projection on N 3D reference coordinate vectors of human face key points in a database in three two-dimensional planes; constructing a joint encoding sub-network based on the k-order improved hourglass network, and jointly encoding the N 2D reference coordinate vectors under each 2D view angle into a 2D joint thermodynamic diagram by utilizing the joint encoding sub-network; superposing the 2D joint thermodynamic diagrams under the three 2Dview angles into a 3D joint thermodynamic diagram by adopting a concat method; and constructing the decoding sub-network based on a 2D full convolution network, and decoding the 3D joint thermodynamic diagram into N 3D detection coordinate vectors by using the decoding sub-network. According to the invention, a corresponding lightweight neural network (a joint coding sub-network and a decoding sub-network) is designed to carry out joint thermodynamic diagram generation and 3D coordinate regression; according to the method, the advantages of the existing 2D and 3D face key point detection method are combined, the model parameter quantity is reduced while the high detection precision is maintained, and the model operation speed is improved.

Description

technical field [0001] The invention relates to the technical fields of image processing and computer vision, in particular to a lightweight human face 3D key point detection method and system. Background technique [0002] With the vigorous development of deep learning technology in the field of computer vision, various face image processing tasks have been widely used in life, among which face key point detection has played an important role in face recognition, expression recognition, face reconstruction, etc. important role. [0003] In the past decade, great achievements have been made in face key point detection, especially in the field of 2D face key point detection. Among them, the ASM (Active Shape Model) algorithm based on the point distribution model proposed by Cootes et al. is a classic face key point detection algorithm. The algorithm first calibrates the training set by manual calibration, obtains the shape model after training, and then passes the key point ...

AI Technical Summary

Problems solved by technology

[0005] However, due to the increase of 3D spatial dimension, the processing speed and model accuracy of the corresponding algorithms are facing huge challenges. The existing 3D face key point detection algorithms have different degrees in terms of processing speed, model size and complexity, and model accuracy. Defects

Images