An end-to-end exemplar association for unsupervised person Re-identification. (September 2020)

• Record Type: Journal Article
• Authors: Wu, Jinlin; Yang, Yang; Lei, Zhen; Wang, Jinqiao; Li, Stan Z.; Tiwari, Prayag; Pandey, Hari Mohan
• Is Part Of: Neural networks. Volume 129(2020)
• Page Start: 43
• Page End: 54
• Publication Date: 2020-09
• Subjects: End-to-end exemplar-based training -- Exemplar association -- Dynamic selection threshold; Neural computers -- Periodicals; Neural networks (Computer science) -- Periodicals; Neural networks (Neurobiology) -- Periodicals; Nervous System -- Periodicals; Ordinateurs neuronaux -- Périodiques; Réseaux neuronaux (Informatique) -- Périodiques; Réseaux neuronaux (Neurobiologie) -- Périodiques; Neural computers; Neural networks (Computer science); Neural networks (Neurobiology); Periodicals; 006.32
• DOI: 10.1016/j.neunet.2020.05.015
• Languages: English
• ISSNs: 0893-6080

Abstract

Abstract: Tracklet association methods learn the cross camera retrieval ability though associating underlying cross camera positive samples, which have proven to be successful in unsupervised person re-identification task. However, most of them use poor-efficiency association strategies which costs long training hours but gains the low performance. To solve this, we propose an effective end-to-end exemplar associations (EEA) framework in this work. EEA mainly adapts three strategies to improve efficiency: (1) end-to-end exemplar-based training, (2) exemplar association and (3) dynamic selection threshold . The first one is to accelerate the training process, while the others aim to improve the tracklet association precision. Compared with existing tracklet associating methods, EEA obviously reduces the training cost and achieves the higher performance. Extensive experiments and ablation studies on seven RE-ID datasets demonstrate the superiority of the proposed EEA over most state-of-the-art unsupervised and domain adaptation RE-ID methods.