Facial Recognition Utilizing the CLM Model and Supervised Machine Learning: A Comparative Analysis

Facial recognition is a common challenge in artificial intelligence. We extensively utilized this program in our daily life. Face recognition software was implemented on many phones to protect user identities and promptly recognize Facebook users in photographs. Despite the numerous methodologies proposed to date, the actual implementation of facial recognition continues to pose significant challenges. Distinguishing individuals by a fundamental method relies on various factors, including lighting, postural differences, and partial facial obstruction. This study aims to develop a facial recognition technique utilizing a constrained local model (CLM) and a machine learning algorithm. It is evaluated with KNN, SVR, and CCA. Furthermore, it has employed CLM and CCA to achieve a maximum recognition accuracy of 3.98 MAE.