Pedestrian re-identification method, system and device and computer readable storage medium

Abstract

The invention discloses a pedestrian re-identification method, system and device, and a computer readable storage medium. The method comprises the steps of obtaining a to-be-trained sample set; basedon the to-be-trained sample set, training a pre-established pedestrian re-identification model through a data re-sampling and cross validation method to obtain a trained pedestrian re-identification model; performing pedestrian re-identification based on the trained pedestrian re-identification model, wherein after the to-be-trained sample set is grouped according to a cross validation method, pedestrian categories between any two groups are different. In the present application, according to the pedestrian re-identification method, the pre-established pedestrian re-identification model is trained through the data re-sampling and cross-validation method, so that the number of the to-be-trained sample sets after data re-sampling is increased, pedestrian categories in each group in the cross-validation method are different, the training effect is enhanced, and the pedestrian re-identification effect can be further improved.

Description

technical field [0001] The present application relates to the technical field of pedestrian re-identification, and more specifically, to a pedestrian re-identification method, system, device, and computer-readable storage medium. Background technique [0002] Person Re-identification (Person Re-identification, Re-ID) is a technology that uses computer vision technology to determine whether a specific pedestrian exists in an image or video sequence. This technology is usually used in the security field. In surveillance video, due to the low resolution of the camera and the special shooting angle, it is usually impossible to obtain high-quality pictures. Therefore, when technologies such as face recognition fail, ReID becomes an important alternative technology. [0003] Pedestrian re-identification is a pedestrian who is interested in a camera, and retrieves all pictures of the pedestrian in other cameras. Data sets are usually pictures obtained by target detection or manu...

AI Technical Summary

Problems solved by technology

Training and testing require that the identity of the person is not repeated. However, in person re-identification, when the size of the dataset is too small, the quality of the dataset is not high, and there is a serious category imbalance problem, how to ensure the recognition effect of person re-identification is easy. become a problem

Images