Non-biological facial signal interception method and system

• Main Authors: WANG CHANGBO, CHEN HUIPING, HU YANPENG, ZHENG ZHIHONG, SHA MAN, LIU SHUBAO, XU JUNFENG, JI QIUYUN, SONG HAICHUAN
• Format: Patent
• Language: Chinese; English
• Published: 17.05.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention discloses a non-biological face signal interception method and system, and the method comprises the steps: carrying out the automatic frame selection of a face region in an image containing a face, and carrying out the depth estimation of a real face through a depth estimation network. The method comprises the following steps: extracting normalized features of image data by using an adaptive normalized algorithm, embedding the adaptive normalized algorithm into a main feature extraction network to carry out cooperative training and extract face features, obtaining classification vectors through a classification network, and obtaining a network model through training; and the obtained network model is tested by using the data of the test set, and the model can be used for judging the type of the face image needing to be deduced after the test is passed. On the basis of the method, a non-biological face signal interception system is designed, the front end of the system is responsible for collecti