Filters








342 Hits in 5.4 sec

Unsupervised Person Re-identification via Softened Similarity Learning [article]

Yutian Lin, Lingxi Xie, Yu Wu, Chenggang Yan, Qi Tian
2020 arXiv   pre-print
Person re-identification (re-ID) is an important topic in computer vision.  ...  computation and a softened classification task.  ...  First, we propose an unsupervised re-ID framework via softened similarity learning.  ... 
arXiv:2004.03547v1 fatcat:7cs6jjfz2jgzxljkisehidm4w4

Unsupervised Clustering Active Learning for Person Re-identification [article]

Wenjing Gao, Minxian Li
2021 arXiv   pre-print
Supervised person re-identification (re-id) approaches require a large amount of pairwise manual labeled data, which is not applicable in most real-world scenarios for re-id deployment.  ...  Towards this goal, we present a Unsupervised Clustering Active Learning (UCAL) re-id deep learning approach.  ...  Unsupervised person re-identification via softened similarity learning. In Proc. IEEE Conf. Comput. Vis.  ... 
arXiv:2112.13308v1 fatcat:kk5d5ghlh5bvvblpsdcirmfow4

Unsupervised Person Re-identification via Multi-order Cross-view Graph Adversarial Network

Xiang Fu, Xinyu Lai
2021 IEEE Access  
INDEX TERMS Unsupervised person re-identification, cross-view graph, graph adversarial network, multi-order correlations.  ...  Unsupervised person re-identification (re-id) is an effective analysis for video surveillance in practice, which can train a pedestrian matching model without any annotations, and it is easy to deploy  ...  improvements for unsupervised cross-view person re-identification problem.  ... 
doi:10.1109/access.2020.3048834 fatcat:v3bhppj4qnccjmsmmrzfgbwie4

Unsupervised Person Re-identification via Multi-Label Prediction and Classification based on Graph-Structural Insight [article]

Jongmin Yu, Hyeontaek Oh
2021 arXiv   pre-print
This paper addresses unsupervised person re-identification (Re-ID) using multi-label prediction and classification based on graph-structural insight.  ...  The proposed GSMLP and SMLC boost the performance of unsupervised person Re-ID without any pre-labelled dataset.  ...  Introduction Person re-identification (Re-ID) have been achieved great success alongside with the development of various deep learning-based methods [13, 8, 23] for extracting discriminative features  ... 
arXiv:2106.08798v1 fatcat:igi5tccyijeorfgp674crcfrby

Unsupervised Vehicle Re-Identification via Self-supervised Metric Learning using Feature Dictionary [article]

Jongmin Yu, Hyeontaek Oh
2021 arXiv   pre-print
The key challenge of unsupervised vehicle re-identification (Re-ID) is learning discriminative features from unlabelled vehicle images.  ...  This paper addresses an unsupervised vehicle Re-ID method, which no need any types of a labelled dataset, through a Self-supervised Metric Learning (SSML) based on a feature dictionary.  ...  Unsupervised Vehicle Re-identification Compared with the vehicle Re-ID with supervision, the studies on unsupervised vehicle Re-ID is in the beginning because it is intractable to learn discriminative  ... 
arXiv:2103.02250v1 fatcat:pzoxdn5jirhmhmbdmkjelwejxa

Fully Unsupervised Person Re-identification viaSelective Contrastive Learning [article]

Bo Pang, Deming Zhai, Junjun Jiang, Xianming Liu
2021 arXiv   pre-print
Person re-identification (ReID) aims at searching the same identity person among images captured by various cameras.  ...  Representation learning plays a critical role in unsupervised person ReID. In this work, we propose a novel selective contrastive learning framework for unsupervised feature learning.  ...  Especially, for video-based person re-identification, RACE [18] firstly adopt anchor sequences to formulate an anchor graph.  ... 
arXiv:2010.07608v2 fatcat:c4zqhhd2wvex7ijmmvqh667a4m

Unsupervised Domain Adaptive Person Re-Identification via Human Learning Imitation [article]

Yang Peng, Ping Liu, Yawei Luo, Pan Zhou, Zichuan Xu, Jingen Liu
2021 arXiv   pre-print
Unsupervised domain adaptive person re-identification has received significant attention due to its high practical value.  ...  The explored three components, collaborate together to constitute a new method for unsupervised domain adaptive person re-identification, which is called Human Learning Imitation framework.  ...  INTRODUCTION Person re-identification (re-ID) [1] is a task to retrieve images of the same person from different image sets.  ... 
arXiv:2111.14014v2 fatcat:hqesvuwe5jcq7o3bwfonh4cc2u

Unsupervised Person Re-Identification with Multi-Label Learning Guided Self-Paced Clustering [article]

Qing Li, Xiaojiang Peng, Yu Qiao, Qi Hao
2021 arXiv   pre-print
Although unsupervised person re-identification (Re-ID) has drawn increasing research attention recently, it remains challenging to learn discriminative features without annotations across disjoint camera  ...  In this paper, we address the unsupervised person Re-ID with a conceptually novel yet simple framework, termed as Multi-label Learning guided self-paced Clustering (MLC).  ...  methods. h Related work In this section, we review the Person Re-Identification (Re-ID) technology in the view of supervised learning, unsupervised domain adaption (UDA), and unsupervised learning.  ... 
arXiv:2103.04580v1 fatcat:rkdapufnwfh27gx4htyxjxo7mu

Graph Neural Network Based Attribute Auxiliary Structured Grouping for Person Re-Identification

Geyu Tang, Xingyu Gao, Zhenyu Chen, Huicai Zhong
2021 IEEE Access  
INTRODUCTION P ERSON re-identification (person re-ID) aims at matching the same pedestrian's image from a database across different cameras [1] .  ...  In the past few years, supervised person re-ID approaches based on deep learning have achieved great success [2] - [5] .  ... 
doi:10.1109/access.2021.3069915 fatcat:xlukop5sdjfjbioycioaynzp6q

CycAs: Self-supervised Cycle Association for Learning Re-identifiable Descriptions [article]

Zhongdao Wang, Jingwei Zhang, Liang Zheng, Yixuan Liu, Yifan Sun, Yali Li, Shengjin Wang
2020 arXiv   pre-print
This paper proposes a self-supervised learning method for the person re-identification (re-ID) problem, where existing unsupervised methods usually rely on pseudo labels, such as those from video tracklets  ...  The goal is to construct a self-supervised pretext task that matches the person re-ID objective.  ...  Self-supervised / unsupervised learning finds critical significance in the person re-identification (re-ID) area, because of the high annotation cost.  ... 
arXiv:2007.07577v1 fatcat:xlczy6vrnjgnboqeqdvemoejoy

Context-Aware Unsupervised Clustering for Person Search [article]

Byeong-Ju Han, Kuhyeun Ko, Jae-Young Sim
2021 arXiv   pre-print
person re-identification methods on the benchmark person search datasets.  ...  In this paper, we first introduce a novel framework of person search that is able to train the network in the absence of the person identity labels, and propose efficient unsupervised clustering methods  ...  We firstly introduce the unsupervised person re-identification methods embedding distinct person features from cropped person images without the person identity labels, and then explain the supervised  ... 
arXiv:2110.01341v1 fatcat:jkpeaf5gljazxppqus5brzusbe

Fully Unsupervised Person Re-Identification via Multiple Pseudo Labels Joint Training

Qing Tang, Ge Cao, Kang-Hyun Jo
2021 IEEE Access  
[22] , and FUL-based unsupervised method Softened Similarity Learning (SSL) [23] explored the self-defined part-based self-similarities, as shown in Fig. 4 (a).  ...  The above two factors make supervised person re-ID and UDA-based unsupervised person re-ID are difficult to meet the requirement of practical industry application.  ... 
doi:10.1109/access.2021.3134181 fatcat:vb4sqiqrq5av3jhnu652dxuknu

Noise Resistible Network for Unsupervised Domain Adaptation on Person Re-Identification

Suian Zhang, Ying Zeng, Haifeng Hu, Shuyu Liu
2021 IEEE Access  
Unsupervised domain adaptation on person re-identification (re-ID), which adapts the model trained on source dataset to the target dataset, has drawn increasing attention over the past few years.  ...  To solve this problem, this paper proposes a novel unsupervised domain adaptation re-ID framework named Noise Resistible Network (NRNet), which mainly consists of two dual-stream networks.  ...  UNSUPERVISED PERSON RE-IDENTIFICATION The unsupervised person re-identification is another manner to make re-ID more practical which is closely related to our work.  ... 
doi:10.1109/access.2021.3071134 fatcat:5rdu4en3qrcozhx5smi3tj4ss4

Meta Clustering Learning for Large-scale Unsupervised Person Re-identification [article]

Xin Jin, Tianyu He, Zhiheng Yin, Xu Shen, Tongliang Liu, Xinchao Wang, Jianqiang Huang, Xian-Sheng Hua, Zhibo Chen
2021 arXiv   pre-print
Unsupervised Person Re-identification (U-ReID) with pseudo labeling recently reaches a competitive performance compared to fully-supervised ReID methods based on modern clustering algorithms.  ...  MCL only pseudo-labels a subset of the entire unlabeled data via clustering to save computing for the first-phase training.  ...  /softened label via pairwise similarity computation.  ... 
arXiv:2111.10032v1 fatcat:cdberenm3bgrplezadavzrxtgi

Person search: New paradigm of person re-identification: A survey and outlook of recent works

Khawar Islam
2020 Image and Vision Computing  
In last few years, deep learning has played unremarkable role for the solution of re-identification problem. Deep learning shows incredible performance in person (re-ID) and search.  ...  This task aims to find a probe person from whole scene which shows great significance in video surveillance field to track lost people, re-identification, and verification of person.  ...  [36] constructed unsupervised approach through asymmetric deep metric learning technique for person (re-ID) in an unsupervised manner.  ... 
doi:10.1016/j.imavis.2020.103970 fatcat:g2zuqww7tbdszkxrc2wkrfno2y
« Previous Showing results 1 — 15 out of 342 results