Face Recognition in Different Scenarios Using Siamese Network
Rohith Mathi, Jaswanth Venkat Mothukuri, Venkata Akhil Pasumarthy, Suja Palaniswamy, R Subramani · 2022 IEEE 3rd Global Conference for Advancement in Technology (GCAT) · 2022
Face detection and face recognition are the most sought applications in image processing and computer vision domains. Face recognition has a wide range of applications, and some of its performance must be improved in unconstrained and surveillance applications. In this paper, face recognition has been performed for a person with different backgrounds and different people around him and in various costumes, which is helpful for surveillance applications and HRI. This work proposes an algorithm to recognize a person by matching similarities between the two faces using Siamese Architecture alongside VGG-Face and MTCNN algorithms. The system gives the similarity score for both images with 90% accuracy.