A face recognition system and method for on-site self-learning

Abstract

The invention discloses a face recognition system and method for on-site self-learning. The system includes a face detector, an autoencoder, a face recognizer, and a face database; Detect the position of the face frame, obtain the face image, and input the face image into the autoencoder; the autoencoder is used to convert the face image into high-dimensional features, and then input the converted image features into face recognition device; a face recognizer, which is used to identify the corresponding face identity from the face database; a face database, which is used to store existing face and identity information. The present invention can automatically and autonomously learn model parameters according to on-site sample data without manual intervention, thereby enhancing the adaptability of the model to scene data and improving the recognition accuracy of the model on a specific target group.

Description

technical field [0001] The invention belongs to the technical field of face recognition, and relates to a face recognition system and method, in particular to a face recognition system and method for on-site autonomous learning. [0002] technical background [0003] Facial recognition refers to technologies that can identify or verify the identity of subjects in images or videos. Face recognition is also one of the most challenging biometric methods when deployed in an unconstrained environment due to the high variability in how face images are presented in the real world. The first face recognition algorithm was born in the early 1970s, and the current traditional approach based on human-designed features and traditional machine learning techniques has recently been replaced by deep neural networks trained with very large datasets. [0004] Convolutional neural network (CNN) is the most commonly used class of deep learning methods for face recognition. Facebook's DeepFace...

Images