Construction method and device of high-precision face shape library, equipment and storage medium

Abstract

The invention discloses a construction method and device of a high-precision face shape library, equipment and a storage medium, and relates to the field of computer vision. The method comprises the steps of obtaining a first facial form base, wherein the first facial form base is a human face shape base obtained through first precision human face data and second precision human face data; constructing a second facial form base through the first facial form base and the first fitting error; fitting the disturbance data through the first facial form base to obtain a second fitting error; fitting the second fitting error through the second facial form base to obtain a third fitting error of the second facial form base; and performing iterative updating on the expanded first fitting error according to the second fitting error and the third fitting error, and constructing a high-precision face shape library according to the second face shape base and the first face shape base obtained wheniteration is ended. Therefore, the data demand for the first-precision face data is reduced, the construction cost of the high-precision face shape library is reduced, and the construction efficiencyof the high-precision face shape library is improved.

Description

technical field [0001] The present application relates to the field of computer vision, in particular to a method, device, equipment and storage medium for constructing a high-precision face shape library. Background technique [0002] The 3DMM (3D Morphable Model, three-dimensional deformable model) library is used to provide a standard face model for three-dimensional (3Dimensions, 3D) face reconstruction. [0003] Taking the Basel Face Model (BFM) as an example, the original face data is collected by a high-precision 3D scanner. After obtaining the original face data, non-rigid registration (non-rigid registration) Named as non-rigid icp or nricp) corresponding to the original face data to register, you can get a face grid similar to the original face data under a specific model, and then perform principal component analysis on the obtained face grid Processing to build a BFM shape library. [0004] In the above technical solution, the BFM shape library needs to collect...

AI Technical Summary

Problems solved by technology

[0004] In the above technical solution, the BFM shape library needs to collect a large amount of high-precision face data, which makes the process of building a 3DMM face shape library more difficult

Images