A collaborative training method for face recognition network and pedestrian re-identification network

Abstract

The invention discloses a collaborative training method of a face recognition network and a pedestrian re-identification network, which adopts a double-network parallel network structure to fuse the features of faces and pedestrians, and use the fused features as pedestrian output features to make it more powerful. Face feature expression ability, and according to the different face image clarity, the face recognition network and the pedestrian re-identification network are trained with different supervision signals. When the face image clarity is low, the pedestrian re-identification network is used for training. The result of the weighted addition of the prediction result and the real label is used as a supervisory signal to guide the face recognition network to train; when the face image is high-resolution, the result of the weighted addition of the prediction result of the face recognition network and the real label As a supervisory signal, it guides the training of the person re-identification network, which not only strengthens the influence of face recognition on the results of person re-identification, but also can use the prediction results of person re-identification to guide the classification of face features when the face image is blurred.

Description

technical field [0001] The invention belongs to the field of deep learning for extracting facial features by a deep neural network, relates to neural network, pattern recognition and other technologies, and in particular relates to a collaborative training method of a face recognition network and a pedestrian re-identification network. Background technique [0002] With the rapid development of my country's safe city construction, it is very important for security and public security criminal investigation business to use many video surveillance cameras to quickly and accurately obtain portrait information in the scene. Fast and effective automatic identity verification is becoming more and more urgent in the field of security. The use of video surveillance for face recognition and pedestrian re-identification has attracted more and more attention from the public security department. [0003] The current face recognition and pedestrian re-identification technology achievemen...

AI Technical Summary

Problems solved by technology

However, face recognition and pedestrian re-identification are usually studied as two independent tasks, ignoring the correlation between face recognition and pedestrian re-identification, and relying only on information such as clothes and height for pedestrian re-identification is not efficient.

Images