Pedestrian re-identification method, device, equipment and medium based on deep learning network

Abstract

The invention discloses a pedestrian re-identification method, device, equipment and medium based on a deep learning network; in this solution, in the network training process, it is first necessary to create an isomorphic training network corresponding to the initial pedestrian re-identification network, the The isomorphic training network has multiple isomorphic branches with the same network structure; moreover, this scheme trains the isomorphic training network through the knowledge collaborative loss function, which can realize information interaction between isomorphic branches during the training process, thereby improving network accuracy. Therefore, this program trains the isomorphic training network through the above operations to obtain more accurate final weight parameters; and, since this program only needs to change the network training process, and does not re-identify pedestrians in the network recognition process The network is complicated, so this solution can maximize the potential of the network and improve network performance without increasing the amount of parameters and calculations.

Description

technical field [0001] The present invention relates to the technical field of image recognition, and more specifically, to a pedestrian re-identification method, device, equipment and medium based on a deep learning network. Background technique [0002] Pedestrian re-identification (Person re-identification, Re-ID) is an important image recognition technology, widely used in public security systems, traffic supervision and other fields. Pedestrian re-identification searches cameras distributed in different locations to determine whether the pedestrians in the field of view of different cameras are the same pedestrian. This technology can be used in scenarios such as criminal suspect search and missing child search. Pedestrian re-identification is mainly achieved through deep learning technology. In recent years, with the continuous development of deep learning technology, the task of pedestrian re-identification has also made great progress. However, at present, in order ...

AI Technical Summary

Problems solved by technology

[0004] 2. Deeper, wider or more complex networks usually increase the amount of calculations, which is not conducive to the application of scenarios with high real-time requirements

For example: for the retrieval and tracking of criminal suspects, a large calculation delay will cause the entire system to miss the best opportunity, which will have a negative impact on system functions

Images