FIELD: computer equipment.
SUBSTANCE: present invention relates to computer engineering. First deep training model is trained on the basis of common images of faces. Second deep training model is trained on the basis of extracted images of faces cut from common images of faces. Detecting the face vitality is performed based on the trained first deep training model to obtain a first prediction estimate and a trained second deep training model in order to obtain a second prediction estimate. Prediction estimation result is generated based on the first prediction estimate and the second prediction estimate, and the prediction evaluation result is compared to the threshold value to determine the face vitality recovery result for the extracted faces images.
EFFECT: technical result is improved accuracy of the result of detecting the facial vitality.
8 cl, 6 dwg
| Title | Year | Author | Number |
|---|---|---|---|
| METHOD FOR DETERMINING FACE AUTHORITY BY SEGMENTATION MASKS | 2021 |
|
RU2758966C1 |
| METHOD, TERMINAL AND SYSTEM FOR BIOMETRIC IDENTIFICATION | 2023 |
|
RU2815689C1 |
| INFORMATION DISPLAY METHOD, DEVICE AND DATA MEDIUM | 2021 |
|
RU2817182C1 |
| SEGMENTATION AND PREDICTION OF TIME PATTERNS OF LOW LEVEL TRAILS | 2020 |
|
RU2783420C1 |
| METHOD, TERMINAL AND SYSTEM FOR BIOMETRIC IDENTIFICATION | 2022 |
|
RU2798179C1 |
| METHOD AND DEVICE FOR AUTONOMOUS AUTHENTICATION OF A PERSON | 2017 |
|
RU2697638C1 |
| IMAGE PROCESSING SYSTEM AND METHOD USING DEEP NEURAL NETWORKS | 2018 |
|
RU2743931C1 |
| SYSTEMS AND METHODS OF ASSESSING THE VIABILITY OF EMBRYOS | 2018 |
|
RU2800079C2 |
| FAST CALCULATION OF CONVOLUTIONAL NEURAL NETWORK | 2018 |
|
RU2722473C1 |
| METHOD FOR ISOLATING AND CLASSIFYING BLOOD CELL TYPES USING DEEP CONVOLUTION NEURAL NETWORKS | 2019 |
|
RU2732895C1 |
