Face Image Transformations for Correct Recognition Problems Solving
Mykola Svirnevskyi, Olexander Barmak, Sergiy Ivaschenko, Iurii V. Krak · 2019 IEEE International Conference on Advanced Trends in Information Theory (ATIT) · 2019
The main problem in face recognition is associated with changes in lighting, head position and facial expression. The article focuses on improving the performance of face recognition algorithms by solving this problem. An approach for face recognition with image conversion is proposed. For this approach, face alignment algorithms through 2D transformations and 3D reconstruction of anthropometric points of the face were developed. Testing of the approach showed that facial recognition quality improved.