A Deep Learning Model to Identify Twins and Look Alike Identification Using Convolutional Neural Network (CNN) and to Compare the Accuracy with SVM Approach
Harini Chandana.S, Senthil Kumar R · ECS Transactions · 2022
Aim: The main aim of this work is to identify twins and look alike using Innovative convolutional neural networks (CNN) and comparing them with SVM. Materials and Methods: Two groups such as Support Vector Machine and convolutional Neural Network were considered. Each algorithm took N=2 samples from the dataset collected and performed two iterations on each algorithm to identify twin faces. Result: The accuracy of identical twin-face detection for biometrics was calculated. It seems Innovative CNN got a significantly better accuracy of (94.01%) when compared with SVM, which has had an accuracy of (83.98%) and having the significance value of (p=0.029). Conclusion: In the work, it could have recognized identical twin faces and differentiate them with different facial features for easy detection to biometrics where Innovative CNN got better accuracy than SVM.