Unsupervised Cross-Dataset Person Re-identification by Transfer Learning of Spatial-Temporal Patterns

Lv, Jianming; Chen, Weihang; Li, Qing; Yang, Can

doi:10.1109/cvpr.2018.00829

Cited by 176 publications

(118 citation statements)

References 25 publications

Supporting

Mentioning

118

Contrasting

Order By: Relevance

“…In addition to exploiting visual information to match pedestrians, there are some methods [10], [11], [12], [13] using the spatial context of the cameras and the tempral stamp of visual frames to constrain the learning of person similarities. In [11], [13], different approaches are explored to use the spatiotemporal constraint to eliminate the irrelevant gallery Fig.…”

Section: Methods Considering Camera Informationmentioning

confidence: 99%

“…images. Lv et al [12] propose an unsupervised incremental learning algorithm to mine the spatio-temporal patterns using the time interval of pedestrians transferring across different cameras. In [10], a unified framework is designed, which uses the spatiotemporal relations to perform the camera network topology inference.…”

Section: Methods Considering Camera Informationmentioning

confidence: 99%

“…To address the above issues, we propose to use the camera identity information contained in the existing person reidentification datasets to help the models separate the foreground human bodies from the background regions. Some existing methods [10], [11], [12], [13] explore the camera network topology and spatiotemporal constraints between cameras to refine the person similarities. In contrast, we exploit the camera identity information to directly train the network arXiv:1910.11547v1 [cs.CV] 25 Oct 2019 to learn background feature representations.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

End-to-End Comparative Attention Networks for Person Re-Identification

Líu¹,

Feng²,

Qi³

et al. 2017

IEEE Trans. on Image Process.

562

317

View full text Add to dashboard Cite

Abstract-Person re-identification across disjoint camera views has been widely applied in video surveillance yet it is still a challenging problem. One of the major challenges lies in the lack of spatial and temporal cues, which makes it difficult to deal with large variations of lighting conditions, viewing angles, body poses and occlusions. Recently, several deep learning based person re-identification approaches have been proposed and achieved remarkable performance. However, most of those approaches extract discriminative features from the whole frame at one glimpse without differentiating various parts of the persons to identify. It is essentially important to examine multiple highly discriminative local regions of the person images in details through multiple glimpses for dealing with the large appearance variance.In this paper, we propose a new soft attention based model, i.e., the end-to-end Comparative Attention Network (CAN), specifically tailored for the task of person re-identification. The end-to-end CAN learns to selectively focus on parts of pairs of person images after taking a few glimpses of them and adaptively comparing their appearance. The CAN model is able to learn which parts of images are relevant for discerning persons and automatically integrates information from different parts to determine whether a pair of images belongs to the same person. In other words, our proposed CAN model simulates the human perception process to verify whether two images are from the same person. Extensive experiments on four benchmark person re-identification datasets, including CUHK01, CHUHK03, Market-1501 and VIPeR, clearly demonstrate that our proposed end-to-end CAN for person re-identification outperforms well established baselines significantly and offer new state-of-the-art performance.

show abstract

Section: Methods Considering Camera Informationmentioning

confidence: 99%

Section: Methods Considering Camera Informationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

End-to-End Comparative Attention Networks for Person Re-Identification

Líu¹,

Feng²,

Qi³

et al. 2017

IEEE Trans. on Image Process.

562

317

View full text Add to dashboard Cite

show abstract

“…Currently, most existing work for person Re-ID mainly focuses on the supervised [5], [6], [7], [8] and unsupervised [9], [10], [11], [12], [13] cases. Although the supervised Lei Comparison between inter-camera and intra-camera images on Market1501 [4].…”

Section: Introductionmentioning

confidence: 99%

Progressive Cross-Camera Soft-Label Learning for Semi-Supervised Person Re-Identification

Wang

Huo

et al. 2020

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

In this paper, we focus on the semi-supervised person re-identification (Re-ID) case, which only has the intracamera (within-camera) labels but not inter-camera (crosscamera) labels. In real-world applications, these intra-camera labels can be readily captured by tracking algorithms or few manual annotations, when compared with cross-camera labels. In this case, it is very difficult to explore the relationships between cross-camera persons in the training stage due to the lack of cross-camera label information. To deal with this issue, we propose a novel Progressive Cross-camera Soft-label Learning (PCSL) framework for the semi-supervised person Re-ID task, which can generate cross-camera soft-labels and utilize them to optimize the network. Concretely, we calculate an affinity matrix based on person-level features and adapt them to produce the similarities between cross-camera persons (i.e., cross-camera soft-labels). To exploit these soft-labels to train the network, we investigate the weighted cross-entropy loss and the weighted triplet loss from the classification and discrimination perspectives, respectively. Particularly, the proposed framework alternately generates progressive cross-camera soft-labels and gradually improves feature representations in the whole learning course. Extensive experiments on five large-scale benchmark datasets show that PCSL significantly outperforms the state-of-the-art unsupervised methods that employ labeled source domains or the images generated by the GAN-based models. Furthermore, the proposed method even has a competitive performance with respect to deep supervised Re-ID methods.

show abstract

“…However, we may suffer performance degradation when directly deploying these trained models to a different real-world scenario [15] due to the non-trivial gap between training data and target domain. On the other hand, it is rather strenuous and impractical to annotate the target data, especially for online surveillance videos of large-scale camera network [16]. To directly make full use of the massive and cheap unlabeled video data, person Re-ID by unsupervised learning, where per camerapair ID labeled training data is no longer required, is gaining increasing popularity [17]- [23].…”

Section: Introductionmentioning

confidence: 99%

Progressive Unsupervised Person Re-Identification by Tracklet Association With Spatio-Temporal Regularization

Xie

Zhou

et al. 2021

IEEE Trans. Multimedia

View full text Add to dashboard Cite

Existing methods for person re-identification (Re-ID) are mostly based on supervised learning which requires numerous manually labeled samples across all camera views for training. Such a paradigm suffers the scalability issue since in real-world Re-ID application, it is difficult to exhaustively label abundant identities over multiple disjoint camera views. To this end, we propose a progressive deep learning method for unsupervised person Re-ID in the wild by Tracklet Association with Spatio-Temporal Regularization (TASTR). In our approach, we first collect tracklet data within each camera by automatic person detection and tracking. Then, an initial Re-ID model is trained based on within-camera triplet construction for person representation learning. After that, based on the person visual feature and spatio-temporal constraint, we associate cross-camera tracklets to generate cross-camera triplets and update the Re-ID model. Lastly, with the refined Re-ID model, better visual feature of person can be extracted, which further promote the association of cross-camera tracklets. The last two steps are iterated multiple times to progressively upgrade the Re-ID model. To facilitate the study, we have collected a new 4K UHD video dataset named Campus-4K with full frames and full spatio-temporal information. 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 DukeMTMC-reID and Campus-4K datasets.

show abstract

Unsupervised Cross-Dataset Person Re-identification by Transfer Learning of Spatial-Temporal Patterns

Cited by 176 publications

References 25 publications

End-to-End Comparative Attention Networks for Person Re-Identification

End-to-End Comparative Attention Networks for Person Re-Identification

Progressive Cross-Camera Soft-Label Learning for Semi-Supervised Person Re-Identification

Progressive Unsupervised Person Re-Identification by Tracklet Association With Spatio-Temporal Regularization

Contact Info

Product

Resources

About