Face Recognition for Soldiers with Bulletproof Helmets Using Generative Networks and Data Augmentation
- Authors
- Choi, Minseo; Shin, Wunjong; Pyo, Heesu; Song, Wootaek; Jung, Byungjun; Shin, Minwoo; Paik, Joonki
- Issue Date
- Jan-2024
- Publisher
- Institute of Electrical and Electronics Engineers Inc.
- Keywords
- Data Augmentation; Face Recognition; Generative Model
- Citation
- 2024 International Conference on Electronics, Information, and Communication, ICEIC 2024
- Journal Title
- 2024 International Conference on Electronics, Information, and Communication, ICEIC 2024
- URI
- https://scholarworks.bwise.kr/cau/handle/2019.sw.cau/73331
- DOI
- 10.1109/ICEIC61013.2024.10457279
- ISSN
- 0000-0000
- Abstract
- Face recognition (FR) is a key technology for a wide range of applications, including privacy-preserving security, education, entertainment, and emotion recognition. However, existing FR methods require a large amount of data to train, and their performance can be degraded by factors such as bulletproof helmets worn by soldiers. This paper proposes a new FR method for soldiers with bulletproof helmets that uses generative networks and data augmentation techniques to effectively increase the amount of training data. The proposed method first generates synthetic face images of soldiers with bulletproof helmets using a generative network. Then, it uses data augmentation techniques to further increase the diversity of the training data. The proposed method is evaluated on both our own dataset of soldiers with bulletproof helmets and a public dataset. The results show that the proposed method outperforms existing FR methods on both datasets, especially in the presence of bulletproof helmets. Overall, the proposed method is a promising new approach for FR for soldiers with bulletproof helmets. It can be used to improve the performance of FR systems in military applications, such as soldier identification and authentication. © 2024 IEEE.
- Files in This Item
-
- Appears in
Collections - Graduate School of Advanced Imaging Sciences, Multimedia and Film > Department of Imaging Science and Arts > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.