Method for reconstructing real-time three-dimensional face data with multiple images

Abstract

The invention discloses a method for reconstructing real-time three-dimensional face data with multiple images. The method includes the following steps: constructing shooting systems of eight cameras; performing multi-camera calibration; performing image matching, to be more specific, calculating global matching points of an ith group of images, and calculating accurate matching points of the ith group of images; calculating three-dimensional face coordinates via the matching points and a projection matrix of the cameras, to be more specific, imaging an object point on the third image of the ith group, and taking an intersection of three projection lines as the object point so as to complete reconstructing of the real-time three-dimensional face data. The face data obtained in the method are global, dense and real-time and can be used directly, and the method can be widely applied to the fields such as the animation industry, computer 3D (three-dimensional) games, human-computer interaction, identity recognition, medical virtual surgery, videophones, face recognition, face expression, age simulation, film and TV advertising production and computer cognition.

Description

technical field [0001] The invention relates to a method for three-dimensional reconstruction of human face, in particular to a method for reconstructing real-time three-dimensional data of human face from multiple images. Background technique [0002] The existing three-dimensional face reconstruction methods mainly include elastic mesh deformation method and structured light scanning method. The technical solutions and defects of these two traditional methods are: [0003] (1) Elastic mesh deformation method: use the information extracted from the continuous video image sequence as the control input of the animation system, track the various feature points of the performer's face and map the texture to the polygonal model; the specific method is to use the reflective ball Paste on the face to track the movement of the face, and obtain the three-dimensional data of the face through mathematical modeling from several key points of the face. The disadvantages of this method ...

AI Technical Summary

Problems solved by technology

The disadvantages of this method are: the lack of realism and texture features of continuous changes in facial muscles; marking on the human face will cause discomfort, and the dependence on marking limits the scope of obtaining geometric information from the marking position; and This method belongs to the method of mathematical modeling, and the obtained 3D data is not the real 3D data of the face.

The disadvantage of this method is that special equipment is required, and no matter it is structured light scanning or laser scanning, it takes a process to obtain face data at a certain moment. time, that is to say, when the facial expression changes too fast, such as fast speaking or fast facial muscle movement, the scanning method cannot obtain real-time facial data, so, in principle, this method cannot realize facial recognition. Dynamic real-time reconstruction of expressions

The disadvantage of this method is that the biggest difficulty is stereo matching, which is not yet mature, so that the modeling accuracy cannot be guaranteed

Images