Pedestrian re-identification method based on knowledge distillation

Abstract

The invention discloses a pedestrian re-identification method based on knowledge distillation, and the method comprises the steps: inputting a pedestrian image training set into a teacher network, andinputting the same data set into a student network; through the synergistic effect of student network transfer, characteristic distillation positions and distance loss functions, carrying out distillation at multiple stages of the whole backbone network at the same time, so that the characteristic output of the student network is continuously close to the characteristics output by the teacher network; minimizing and updating parameters of the student model through a distillation loss function, and training a student network; carrying out distance measurement on the obtained feature vectors, searching out a pedestrian target graph with the highest similarity, and finally enabling the accuracy of the student network resnet18 to be greatly improved to be close to the accuracy of the teachernetwork resnet50. According to the method, personnel re-identification is realized by using a knowledge distillation transfer learning method, and the thought of replacing a large model with a small model is adopted, and therefore, calculation complexity can be effectively reduced, and the accuracy of a student model can be ensured.

Description

technical field [0001] The invention relates to the fields of computer vision and image processing, in particular to a pedestrian re-identification method based on knowledge distillation. Background technique [0002] The goal of person re-identification is to find a specific pedestrian in a library of images captured by many different cameras. The difficulty of this problem lies in the following aspects: the shooting angle, pedestrian posture, light intensity and occlusion of different pictures may be quite different. In the pedestrian re-identification module, the specified query image is compared with the pictures in the gallery, and the picture of the same person as the query picture is retrieved. To compare the image in the gallery with the query image, the system first uses hand-crafted descriptors or deep neural networks to extract feature representations describing each image. Usually, the features of the gallery are calculated and stored offline in advance, so at ...

AI Technical Summary

Problems solved by technology

However, the calculation overhead of deep neural network operation is relatively large, so we can adopt the pedestrian re-identification method based on deep learning, and reduce the calculation overhead on the existing deep learning method to better meet the needs of actual scenarios

Images