Three-dimensional face modeling method and three-dimensional face modeling system assisted by face geometric information

Abstract

The invention discloses a three-dimensional face modeling method and a three-dimensional face modeling system assisted by face geometric information. According to the invention, based on a binocular stereoscopic vision principle of active projection, marking the surface of the target face by using phase information contained in a multi-frame stripe structured light place; acquiring a modeling image by using a camera, acquiring relative phase information contained in a stripe structured light place based on a spatial phase unwrapping technology, unifying the relative phase information of a binocular camera to the same reference by taking a human face feature point as an anchor point, performing phase matching to obtain a disparity map, and further reconstructing a three-dimensional human face model. According to the method, the position information of the human face feature points contained in the human face fringe pattern is fully utilized, phase unwrapping can be carried out without adding an additional structure light field, the number of projection and shooting structure light fields needed for completing three-dimensional modeling is effectively reduced, the modeling image collection time is shortened, and the modeling precision is effectively guaranteed.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a three-dimensional face modeling method and system assisted by face geometric information. Background technique [0002] Using digital images to reconstruct the three-dimensional surface shape of objects is a long-term research focus of computer vision. In recent years, 3D face modeling technology has developed rapidly. Realistic 3D face models have broad application prospects in many fields such as film and television, games, education, virtual reality, plastic surgery, identity authentication, and 3D printing. In recent years, public safety and information security have become important areas related to the national economy and the people's livelihood. Biometric recognition technology has developed rapidly in recent years. The recognition rate of recognition technology is easily affected by factors such as posture, lighting, and makeup, and the anti-counterfeiting abi...

AI Technical Summary

Problems solved by technology

The current situation that the temporal phase unwrapping method needs to increase the number of modeling images is not conducive to three-face measurement

In the case of keeping the acquisition time of modeling images unchanged, this contradiction can also be alleviated by improving hardware performance, but this means an increase in cost, which is not conducive to large-scale promotion and application

Images