A craniofacial restoration method based on a deep generative adversarial network

Abstract

The invention discloses a craniofacial restoration method based on a deep generative adversarial network. The craniofacial restoration method comprises the following steps: S1, obtaining a training sample; Obtaining condition information of the to-be-restored object; S2, constructing a generative adversarial network, and obtaining a cost function; Obtaining two-dimensional skull data according tothe three-dimensional skull information of the to-be-restored object; S3, taking data in the training sample as input of the generative adversarial network, training the generative adversarial networkto optimize a cost function, and completing training of the generative adversarial network; And S4, taking the two-dimensional skull data and the condition information of the to-be-restored object asthe input of the trained generative adversarial network to obtain restored craniofacial information. The method is based on the generative adversarial network, high-precision craniofacial restorationcan be carried out according to the skull data and the condition information of the to-be-restored object, and facial recognition of a victim in a criminal case, restoration of the facial appearanceof an ancient person in archaeology, effect prediction of a medical facial surgery and the like are facilitated.

Description

technical field [0001] The invention relates to the field of craniofacial restoration, in particular to a craniofacial restoration method based on a deep confrontation network. Background technique [0002] Based on the skull samples to restore a close to the real craniofacial, for the facial recognition of victims in criminal cases, the restoration of ancient faces in archaeology, the prediction of the effect of medical facial surgery, and the animation production close to real human faces, etc. has a very important role. Because the structure of the human skull is very complex, the inside of the skull contains the brain, blood vessels and nerves, and the outside is covered with multiple layers of soft tissue structure. What is even more difficult is that due to differences in race, fat and thinness, the thickness of each layer of soft tissue is quite different, which makes the facial structure complex and changeable. [0003] Most early craniofacial reconstruction work w...

AI Technical Summary

Problems solved by technology

[0005] In view of the above-mentioned deficiencies in the prior art, a craniofacial restoration method based on a deep generative confrontation network provided by the present invention solves the problem of poor restoration effect of the existing craniofacial restoration methods

Images