Unsupervised Person Re-Identification by Soft Multilabel Learning

@article{Yu2019UnsupervisedPR,
  title={Unsupervised Person Re-Identification by Soft Multilabel Learning},
  author={Hong-Xing Yu and Wei-Shi Zheng and Ancong Wu and Xiaowei Guo and Shaogang Gong and Jianhuang Lai},
  journal={2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2019},
  pages={2143-2152}
}
  • Hong-Xing Yu, W. Zheng, +3 authors J. Lai
  • Published 15 March 2019
  • Computer Science
  • 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Although unsupervised person re-identification (RE-ID) has drawn increasing research attentions due to its potential to address the scalability problem of supervised RE-ID models, it is very challenging to learn discriminative information in the absence of pairwise labels across disjoint camera views. [...] Key Method To enable effecient soft multilabel learning, we introduce the reference agent learning to represent each reference person by a reference agent in a joint embedding. We evaluate our unified deep…Expand
A Deep Clustering-Guide Learning for Unsupervised Person Re-identification
TLDR
A deep clustering-guided model for unsupervised RE-ID that focuses on full mining of supervisions and a complete usage of the mined information, andTrain the model with a reliable and hard mining in both intra-cluster and inter-clusters to learn discriminative features for unlabeled persons. Expand
Weakly Supervised Tracklet Person Re-Identification by Deep Feature-wise Mutual Learning
TLDR
This work proposes a weakly supervised tracklets Re-ID model based on Weakly Supervised Tracklets data from various camera views, which can be inexpensively acquired by combining the fragmented tracklets of the same person in the same camera view over a period of time. Expand
Joint patch and instance discrimination learning for unsupervised person re-identification
TLDR
Extensive experiments on Market-1501 and DukeMTMC-reID datasets demonstrate the effectiveness of the proposed joint patch and instance discrimination learning (JPIL) framework for unsupervised person re-ID. Expand
Hard-sample Guided Hybrid Contrast Learning for Unsupervised Person Re-Identification
  • Zheng Hu, Chuang Zhu, Gang He
  • Computer Science
  • ArXiv
  • 2021
TLDR
This paper proposes a Hard-sample Guided Hybrid Contrast Learning (HHCL) approach combining cluster- level loss with instance-level loss for unsupervised person Re-ID, and applies cluster centroid contrastive loss to ensure that the network is updated in a more stable way. Expand
Distilled Camera-Aware Self Training for Semi-Supervised Person Re-Identification
TLDR
A Distilled Camera-Aware Self Training framework for semi-supervised person re-identification, and a Multi-Teacher Selective Similarity Distillation Loss to selectively aggregate the knowledge of multiple weak teacher models trained with different subsets and distill a stronger student model for self training. Expand
Angular regularization for unsupervised domain adaption on person re-identification
TLDR
This paper aims to address the UDA problem by designing a Strong Clustering-based Unsupervised Re-ID (SCURID) baseline and designing the angular margin losses consisting of Hard Angular Margin Identification (HAMI) loss and Softangular Margin identification (SAMI) loss, which enables the clustering- based UDA Re- ID model to alleviate the negative effect of noisy labels and toward more discriminability. Expand
Unsupervised Person Re-identification via Simultaneous Clustering and Consistency Learning
TLDR
This work designs a pretext task for unsupervised re-identification by learning visual consistency from still images and temporal consistency during training process, such that the clustering network can separate the images into semantic clusters automatically. Expand
Joint Visual and Temporal Consistency for Unsupervised Domain Adaptive Person Re-Identification
TLDR
This paper tackles the challenge of jointly enforcing visual and temporal consistency in the combination of a local one-hot classification and a global multi-class classification by outperforming recent unsupervised domain adaptive methods, which leverage more labels for training. Expand
Progressive Unsupervised Person Re-Identification by Tracklet Association With Spatio-Temporal Regularization
TLDR
Experimental results show that with the spatio-temporal constraint in the training phase, the proposed approach outperforms the state-of-the-art unsupervised methods by notable margins on DukeMTMC-reID, and achieves competitive performance to fully supervised methods on both DukeMT MC-re ID and Campus4K datasets. Expand
Large-Scale Unsupervised Person Re-Identification with Contrastive Learning
TLDR
This work proposes to learn ReID representation from large-scale unlabeled surveillance video alone using contrastive learning and shows far superior performance among unsupervised methods that do not use any training data in the testing domain. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 73 REFERENCES
Unsupervised Person Re-identification by Deep Learning Tracklet Association
TLDR
A Tracklet Association Unsupervised Deep Learning (TAUDL) framework characterised by jointly learning per-camera (within-camera) tracklet association (labelling) and cross-camera tracklet correlation by maximising the discovery of most likely tracklet relationships across camera views. Expand
Unsupervised Person Re-Identification by Deep Asymmetric Metric Embedding
TLDR
A novel unsupervised deep framework named the DEep Clustering-based Asymmetric MEtric Learning (DECAMEL) is developed, which learns a compact cross-view cluster structure of Re-ID data to help alleviate the view-specific bias and facilitate mining the potential cross-View discriminative information for unsuper supervised Re- ID. Expand
Transferable Joint Attribute-Identity Deep Learning for Unsupervised Person Re-identification
TLDR
This work introduces an Transferable Joint Attribute-Identity Deep Learning (TJ-AIDL) for simultaneously learning an attribute-semantic and identity-discriminative feature representation space transferrable to any new (unseen) target domain for re-id tasks without the need for collecting new labelled training data from the target domain. Expand
Cross-View Asymmetric Metric Learning for Unsupervised Person Re-Identification
TLDR
The model aims to learn an asymmetric metric, i.e., specific projection for each view, based on asymmetric clustering on cross-view person images, and finds a shared space where view-specific bias is alleviated and thus better matching performance can be achieved. Expand
Person Re-identification by Unsupervised ` 1 Graph Learning
Most existing person re-identification (Re-ID) methods are based on supervised learning of a discriminative distance metric. They thus require a large amount of labelled training image pairs whichExpand
Deep Association Learning for Unsupervised Video Person Re-identification
TLDR
A novel Deep Association Learning (DAL) scheme is formulated, the first end-to-end deep learning method using none of the identity labels in model initialisation and training, which significantly outperforms current state-of-the-art unsupervised video person re-id methods on three benchmarks. Expand
Dictionary Learning with Iterative Laplacian Regularisation for Unsupervised Person Re-identification
TLDR
A new dictionary learning for sparse coding formulation with a graph Laplacian regularisation term whose value is set iteratively enables the exploitation of cross-view identity-discriminative information ignored by existing unsupervised Re-ID methods. Expand
Person Re-Identification by Iterative Re-Weighted Sparse Ranking
TLDR
The approach makes use of soft- and hard- re-weighting to redistribute energy among the most relevant contributing elements and to ensure that the best candidates are ranked at each iteration of an iterative extension to sparse discriminative classifiers capable of ranking many candidate targets. Expand
A Pose-Sensitive Embedding for Person Re-identification with Expanded Cross Neighborhood Re-ranking
TLDR
This paper proposes an effective approach that incorporates both the fine and coarse pose information of the person to learn a discriminative embedding and shows that both the learned representation and the re-ranking method achieve state-of-the-art performance on a number of challenging surveillance image and video datasets. Expand
Cross-Scenario Transfer Person Reidentification
TLDR
The proposed method, called the constrained asymmetric multitask discriminant component analysis (cAMT-DCA), can be solved by a simple eigen decomposition with a closed form, getting rid of any iterative learning used in most conventional MTL analyses. Expand
...
1
2
3
4
5
...