Unsupervised Domain Adaptation in Person Reidentification by Learning the Features of Both Source and Target Domains

Document Type : Research Paper

Authors

1 MS in Information Technology Engineering, Deep Learning Research Lab, Department of Computer Engineering, College of Farabi, University of Tehran

2 Deep Learning Research Lab, Department of Computer Engineering, College of Farabi, University of Tehran

3 Department of Computer Engineering, College of Farabi, University of Tehran

Abstract

Person reidentification problem is intended to retrieve images of one person from the images captured by non-overlapping cameras. Despite the successful performance of the deep person reidentification models, the performance usually decreases during testing the model on different unlabeled datasets.
In this paper, a well-generalized model for unsupervised domain adaptation in person reidentificationis proposed. The model uses both labeled source dataset and unlabeled target dataset during training and the goal is to generalize well on the unlabeled target domain. To this end, our model is optimized by three loss functions. The final loss function consists of one loss function for supervised learning of the source domain’s features, another for unsupervised learning of the target domain’s features, and a triplet loss function for learning the features of both source and target domains. The proposed model with strategy 2 for selecting neighbors achieves 84.5 % in rank-1 accuracy and 63% for mAP on Duke -> Market setting. It also achieves 70.1 % in rank-1 accuracy and 49.1 % for mAP on Market -> Duke setting.
 

Keywords

Journal of Machine Vision and Image Processing
Volume 10, Issue 1
March 2023
Pages 1-15

Files

Share

How to cite

Statistics