Data-augmented pedestrian re-identification method based on generative adversarial network model

Abstract

The invention relates to a data-enhanced pedestrian re-identification method based on a generative confrontation network model, which includes using the Mask-RCNN image segmentation algorithm to segment a mask image of a pedestrian in an image; combining the mask image and manually marking pedestrian attributes, training a terminal An end-to-end improved star-shaped generative adversarial network generates fake training images under any number of cameras from real pedestrian images under one camera; uses the trained improved star-shaped generative adversarial network to generate fake images of all camera domains corresponding to all real images Training images; send real images and fake training images together to the pedestrian re-identification model, calculate the distance between pedestrian images and complete the pedestrian re-identification function. The invention has a reasonable design, uses the generative confrontation network to generate more training samples, and the generated image background can effectively represent the real scene under the corresponding camera, effectively improves the robustness and judgment ability of the pedestrian re-identification model, and effectively improves the the accuracy of pedestrian re-identification.

Description

technical field [0001] The invention belongs to the technical field of computer vision, in particular to a data-enhanced pedestrian re-identification method based on a generative confrontation network model. Background technique [0002] In recent years, security camera networks have spread all over the city, and with it comes the storage of massive video data. However, due to the time-consuming and laborious manual analysis of a large amount of video content, person re-identification technology, as an important part of intelligent video surveillance, has attracted the attention of researchers all over the world. [0003] Pedestrian re-identification technology refers to the specific pedestrian technology that uses computer vision technology to match different scenes in a non-overlapping multi-camera system. Due to the change of the environment where the camera is located and the non-rigid motion of pedestrians, such as brightness changes, pose changes, occlusion, and backg...

AI Technical Summary

Problems solved by technology

However, existing deep learning models still have shortcomings: in actual scenes, the number of images of pedestrians is extremely large, and the scene changes are diverse. However, the size of the existing public datasets is relatively small , and only collect images of pedestrians under a few specific cameras

In addition, due to the relatively small number of cameras used in the collection of existing data sets, the internal environment changes are relatively fixed, and it is difficult to match the real scene

Therefore, the existing deep learning-based pedestrian re-identification models are prone to overfitting in the training process using public datasets.

Images