.png){kind=link}
This project presents a novel person identification framework for the COVID-19 and post-COVID-19 era. The motivation of the proposed framework is based on the challenges being faced by person identification systems especially employing face recognition algorithms. Due to the ongoing pandemic, face-masks or face coverings have become a new normality which hinders effective identification by state of the art face recognition systems since more than 50% of the face features are unavailable causing drop in accuracy of identification system. The premise of the proposed framework is to improve the accuracy of existing face recognition systems by exploring complementary biometrics which could effectively counter the drop in accuracy of the face recognition systems. The choice of the biometric is guided by contact-less identification due to the pandemic for which the human GAIT is the best bet. The proposed framework exploits the GAIT biometric as an additional modality which when combined with masked faces via a novel fusion strategy is able to achieve better overall person identification accuracy as opposed to the individual modalities. Extensive experimental validation is carried out to establish the proposed bi-modal person identification system which could potentially be a boon during the current pandemic as well as in the future when contact-less biometrics becomes a new normality.
One can check out the following links to know about the data set preparation strategy for Face 😷 and Gait 🚶 Biometrics:
