Answers are grounded in this paper and cite their sources.
Face recognition systems have gained significant attention in recent years due to their applications in various domains such as security, authentication and surveillance. However, the performance of these systems heavily relies on the quality of input face images. This paper presents a quality assessment technique for face images with the aim of enhancing the accuracy and reliability of face recognition systems. There are several methods used for Face Image Quality Assessment (FIQA), thus measuring the quality of images is likely to filter out poor-quality images, which can improve the performance of downstream tasks. Therefore, a novel FIQA framework is proposed to select the best facial image for better recognition. In our proposed methodology, firstly measures various image quality factors then apply fusion techniques to combine all quality factors for classifying the face quality. In addition, conducting a statistical test to check the level of confidence of proposed model. The technique is evaluated on the Color-Feret and KinectFace datasets. The correlation determine between the proposed system and human score is 0.95.
No takes yet. Share an insight, caveat, or question.
Phalguni Gupta Parul Choudhary (2023) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: