Periocular Recognition

Periocular recognition, focusing on the area around the eyes, aims to develop robust and accurate biometric identification systems. Current research heavily utilizes deep learning, employing convolutional neural networks (CNNs) and vision transformers (ViTs), often in combination, to extract features from periocular images for recognition and also leverages these models for segmentation and distance prediction to aid in disease diagnosis and monitoring. This field is significant due to its potential for reliable authentication in various applications, particularly where other biometric methods are unreliable, such as in scenarios with face occlusion or poor lighting conditions, and its growing use in ophthalmic applications.