A three-dimensional texture photography system, calibration method and imaging method

Abstract

The invention relates to a three-dimensional texture information measurement technology in the field of three-dimensional face recognition, in particular to a three-dimensional texture photography system, a calibration method and an imaging method. The invention proposes the concept of uncoupled three-dimensional texture information acquisition for the first time. In the invention, multi-measurement units are used to obtain texture images from different angles of view. The measured normal of the complete three-dimensional face triangle slice determines the difference between two-dimensional texture information and three-dimensional texture information at different angles. The constructed weights can realize 3D full face shape data and 3D color texture acquisition in the sense of uncoupled or low-coupled real space at one time. Provide a reliable method of taking pictures of faces for 3D face recognition technology.

Description

technical field [0001] The invention relates to the fields of three-dimensional face recognition and three-dimensional measurement, in particular to a three-dimensional texture photography system, a calibration method and an imaging method. Background technique [0002] In recent years, due to the needs of aerospace security, homeland security, social security, information security and mobile payment, all countries in the world have increased investment in the security field, and fast, accurate and natural identification is exactly what the security field is facing. A core and urgent problem to be solved. Although face recognition based on two-dimensional images has made great progress and can achieve good recognition performance in constrained environments, it is still greatly affected by changes such as illumination, posture, and expression. Compared with 2D face recognition, 3D face recognition not only uses the 3D shape information of the curved surface of the face, but...

AI Technical Summary

Problems solved by technology

There are already 3D products on the market, such as Microsoft's Kinect and Intel's Realsense, which can achieve real-time acquisition of depth data, but the accuracy cannot be guaranteed, and the obtained is two-dimensional texture information

Neither the published scientific research results nor the products mentioned above can realize the synchronous photography of 3D texture information, which greatly limits the advantages of 3D face recognition not being restricted by poses.

Images