Semi-supervised character re-identification method based on camera style and human body posture adaptation

Abstract

The invention discloses a semi-supervised character re-identification method based on camera styles and human body posture adaptation. The method comprises the following steps: selecting labeled datacovering all identities and all camera styles; carrying out data expansion on the labeled data to generate images with different postures in the camera and images with different styles between the cameras; training a network model by combining the expanded labeled data with the unlabeled data; and performing pseudo label distribution by a strategy of constraining the average distance between the label-free data and the label-containing data characteristics and the number of each type of images. According to the method, the problems of small label data volume and unreasonable pseudo label distribution in the prior art are solved.

Description

technical field [0001] The invention belongs to the technical field of deep learning and machine vision, and in particular relates to a semi-supervised person re-recognition method based on camera style and human body posture adaptation. Background technique [0002] Person re-identification is the purpose of cross-camera view retrieval and detection of pedestrian images in the field of machine vision. Traditional person re-identification methods usually require the help of pedestrian identity information, that is, supervised person re-identification, but obtaining annotation information has disadvantages such as poor implementability and high labor costs. , and in order to make full use of these limited annotation information, traditional person re-identification methods usually focus on the depth of the network layer and the complexity of the network structure, which will obviously lead to a substantial increase in computing and storage requirements. Therefore, in recent y...

AI Technical Summary

Problems solved by technology

[0007] Aiming at the deficiencies in the prior art, the present invention provides a semi-supervised person re-identification method based on camera style and human body posture adaptation. The person re-identification task is greatly affected by factors such as camera parameters, shooting angle, image resolution, and human body posture. To solve the problem, optimize the labeled data expansion and pseudo-label estimation to form intra-camera pose learning and inter-camera style learning, get more labeled data with different poses and styles, and redesign the pseudo-label allocation strategy. Improve the performance of person re-identification; solve the problem of low amount of labeled data and unreasonable distribution of pseudo-labels in the prior art

Images