P-PCC: Parallel pearson correlation condition for robust cosmetic makeup face recognitions

Abstract

The performance of face recognition has been improved over the past; however, there remain some limitations, especially with noises and defects, such as occlusion, face pose, expression, and,in particular, cosmetic makeup change. Recently, the makeup has directly impacted on face characteristics, e.g., face shape, texture, and color, perhaps leading to low classification precision. Thus, this research proposes a robust approach to enhance the recognition accuracy for the makeup using Pearson Correlation (PC) combining with the channel selection (PCC). To further optimize the complexity, the parallelism of PCC was then investigated. This technique demonstrates the practicality and proficiency by outperforming the accuracy and computational time over a traditional PCA and PC.