Fully Unsupervised Person Re-Identification by Enhancing Cluster Samples

doi:10.1145/3507971.3507984

CORC > 西安光学精密机械研究所 > 中国科学院西安光学精密机械研究所 > 光学影像学习与分析中心

	Fully Unsupervised Person Re-Identification by Enhancing Cluster Samples
	Chen, Xiumei 2,3; Zheng, Xiangtao 3; Zhu, Kaijian 1; Lu, Xiaoqiang 3
	2021-12-16
会议日期	2021-12-16
会议地点	Virtual, Online, China
关键词	unsupervised person re-identification sample transformation convolutional neural network feature extraction
DOI	10.1145/3507971.3507984
页码	70-73
英文摘要	Fully unsupervised person re-identification aims to train a discriminative model with unlabeled person images. Most existing methods first generate pseudo labels by clustering image features (convolutional features) and then fine-tune the convolutional neural network (CNN) with pseudo labels. However, these methods are greatly limited by the quality of the pseudo labels. In this paper, a cluster sample enhancement method is introduced to increase the reliability of pseudo-label samples to facilitate the CNN training. Specifically, when generating pseudo labels, only the samples with high-confidence pseudo-label predictions are selected. In addition, to enhance the selected samples for training, two different image transformations are adopted and coupled with specific-design loss functions to boost the model performance. Experiments demonstrate the effectiveness of the proposed method. Concretely, the proposed method achieves 87.1% rank-1 and 70.2% mAP accuracy on Market-1501. © 2021 ACM.
产权排序	1
会议录	2021 7th International Conference on Communication and Information Processing, ICCIP 2021
会议录出版者	Association for Computing Machinery
语种	英语
ISBN号	9781450385190
内容类型	会议论文
源URL	[http://ir.opt.ac.cn/handle/181661/95804]
专题	西安光学精密机械研究所_光学影像学习与分析中心
作者单位	1.State Key Laboratory of Astronautic Dynamics, Xi'an Satellite Control Center, Xi'an, Shaanxi, China 2.University of Chinese Academy of Sciences Beijing, China; 3.Key Laboratory of Spectral Imaging Technology CAS, Xi'an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences, Xi'an, Shaanxi, China;
推荐引用方式 GB/T 7714	Chen, Xiumei,Zheng, Xiangtao,Zhu, Kaijian,et al. Fully Unsupervised Person Re-Identification by Enhancing Cluster Samples[C]. 见:. Virtual, Online, China. 2021-12-16.