To obtain access to full text of journal and articles you must register!
- Article name
- Methods for determining the live presence of a user in front of a video camera in the task of biometric face authentication
- Authors
- Sulavko A. E., , sulavich@mail.ru, Omsk State Technical University, Omsk, Russia
Panfilova I. E., , panfilova_2015@bk.ru, Samara State Technical University, Samara, Russia
- Keywords
- spoofing attacks / face recognition / biometric authentication / deep learning / convolutional neural network / computer vision
- Year
- 2023 Issue 2 Pages 17 - 26
- Code EDN
- ENKYIW
- Code DOI
- 10.52190/2073-2600_2023_2_17
- Abstract
- This paper provides a general overview of the methods and technologies of liveness detection. Among the results of the analysis, we can single out a steady trend of changing approaches to determining live presence based on "manual" processing of images to multilayer machine learning algorithms. However, the analysis of such algorithms shows that a characteristic feature of their functioning is the impossibility of reproducing the results obtained during training in real practice. Moreover, even minor changes in the conditions of the authentication procedure for such algorithms become critical in terms of the robustness of the entire system. A possible solution to these problems can be the use of explainable artificial intelligence methods for the tasks of liveness detection.
- Text
- BUY for read the full text of article
- Buy
- 500.00 rub
