Person re-identification by unsupervised video matching

Ma, Xiaolong; Zhu, Xiatian; Gong, Shaogang; Xie, Xudong; Hu, Jianming; Lam, Kin‐Man; Zhong, Yisheng

doi:10.1016/j.patcog.2016.11.018

Cited by 106 publications

(58 citation statements)

References 90 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hence, their scalability and usability is poor for real-world re-id deployments where no such large training sets are available for every camera pair. Classical unsupervised learning methods based on hand-crafted features offer poor re-id performance [14,22,24,25,32,35,37,47,49,55,59] when compared to the supervised learning based re-id models. While a balancing trade-off between model scalability and re-id accuracy can be achieved by semi-supervised learning [33,49], these models still assume sufficiently large sized cross-view pairwise labelled data for model training.…”

Section: Related Workmentioning

confidence: 99%

Unsupervised Person Re-identification by Deep Learning Tracklet Association

Zhu²,

Gong

2018

Lecture Notes in Computer Science

Self Cite

218

214

View full text Add to dashboard Cite

Most existing person re-identification (re-id) methods rely on supervised model learning on per-camera-pair manually labelled pairwise training data. This leads to poor scalability in practical re-id deployment due to the lack of exhaustive identity labelling of image positive and negative pairs for every camera pair. In this work, we address this problem by proposing an unsupervised re-id deep learning approach capable of incrementally discovering and exploiting the underlying re-id discriminative information from automatically generated person tracklet data from videos in an end-to-end model optimisation. We formulate 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. Extensive experiments demonstrate the superiority of the proposed TAUDL model over the state-of-the-art unsupervised and domain adaptation reid methods using six person re-id benchmarking datasets.

show abstract

Section: Related Workmentioning

confidence: 99%

Unsupervised Person Re-identification by Deep Learning Tracklet Association

Zhu²,

Gong

2018

Lecture Notes in Computer Science

Self Cite

218

214

View full text Add to dashboard Cite

show abstract

“…In this section, we compare our method with state-of-the-art approaches. Recent works [25], [26] show that spatiotemporal features from sequences can be combined with distance metric learning algorithms to achieve better performance. Thus, we combine the learned deep features with two supervised metric learning methods: Local Fisher Discriminant Analysis (LFDA [4]) and KISSME [38].…”

Section: G Comparison With State-of-the-art Approachesmentioning

confidence: 99%

Where-and-When to Look: Deep Siamese Attention Networks for Video-Based Person Re-Identification

Wang

Gao

et al. 2019

IEEE Trans. Multimedia

218

111

View full text Add to dashboard Cite

Video-based person re-identification (re-id) is a central application in surveillance systems with significant concern in security. Matching persons across disjoint camera views in their video fragments is inherently challenging due to the large visual variations and uncontrolled frame rates. There are two steps crucial to person re-id, namely discriminative feature learning and metric learning. However, existing approaches consider the two steps independently, and they do not make full use of the temporal and spatial information in videos. In this paper, we propose a Siamese attention architecture that jointly learns spatiotemporal video representations and their similarity metrics. The network extracts local convolutional features from regions of each frame, and enhance their discriminative capability by focusing on distinct regions when measuring the similarity with another pedestrian video. The attention mechanism is embedded into spatial gated recurrent units to selectively propagate relevant features and memorize their spatial dependencies through the network. The model essentially learns which parts (where) from which frames (when) are relevant and distinctive for matching persons and attaches higher importance therein. The proposed Siamese model is end-to-end trainable to jointly learn comparable hidden representations for paired pedestrian videos and their similarity value. Extensive experiments on three benchmark datasets show the effectiveness of each component of the proposed deep network while outperforming state-of-the-art methods.

show abstract

“…Liang et al [38] adopt probabilistic model to organize and depict the spatial feature distribution of person images, which is a robust approach against environment changes and external interference. Ma et al [50] proposed a novel video based person Re-Id approach to match pedestrian across views. A novel space-time person representation in form of sequence is generated based on existing action space-time features and spatio-temporal pyramids.…”

Section: Person Re-id Via Unsupervised Learningmentioning

confidence: 99%

PAC-GAN: An effective pose augmentation scheme for unsupervised cross-view person re-identification

et al. 2020

View full text Add to dashboard Cite

Person re-identification (person Re-Id) aims to retrieve the pedestrian images of a same person that captured by disjoint and non-overlapping cameras. Lots of researchers recently focuse on this hot issue and propose deep learning based methods to enhance the recognition rate in a supervised or unsupervised manner. However, two limitations that cannot be ignored: firstly, compared with other image retrieval benchmarks, the size of existing person Re-Id datasets are far from meeting the requirement, which cannot provide sufficient pedestrian samples for the training of deep model; secondly, the samples in existing datasets do not have sufficient human motions or postures coverage to provide more priori knowledges for learning. In this paper, we introduce a novel unsupervised pose augmentation cross-view person Re-Id scheme called PAC-GAN to overcome these limitations. We firstly present the formal definition of cross-view pose augmentation and then propose the framework of PAC-GAN that is a novel conditional generative adversarial network (CGAN) based approach to improve the performance of unsupervised corss-view person Re-Id. Specifically, The pose generation model in PAC-GAN called CPG-Net is to generate enough quantity of pose-rich samples from original image and skeleton samples. The pose augmentation dataset is produced by combining the synthesized poserich samples with the original samples, which is fed into the corss-view person Re-Id model named Cross-GAN. Besides, we use weight-sharing strategy in the CPG-Net to improve the quality of new generated samples. To the best of our knowledge, we are the first try to enhance the unsupervised cross-view person Re-Id by pose augmentation, and the results of extensive experiments show that the proposed scheme can combat the state-of-the-arts.

show abstract

Person re-identification by unsupervised video matching

Cited by 106 publications

References 90 publications

Unsupervised Person Re-identification by Deep Learning Tracklet Association

Unsupervised Person Re-identification by Deep Learning Tracklet Association

Where-and-When to Look: Deep Siamese Attention Networks for Video-Based Person Re-Identification

PAC-GAN: An effective pose augmentation scheme for unsupervised cross-view person re-identification

Contact Info

Product

Resources

About