Instance-aware diversity feature generation for unsupervised person re-identification

Unsupervised person re-identification (Re-ID) methods have made significant progress by exploiting contrastive learning from unlabeled data. However, the previous approaches including cluster-level or instance-level contrast loss, did not fully explore inherent commonality of each identified individ...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Displays Ročník 83; s. 102717
Hlavní autori: Zhang, Xiaowei, Dou, Xiao, Zhao, Xinpeng, Li, Guocong, Wang, Zekang
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Elsevier B.V 01.07.2024
Predmet:
Contrastive learning
Diversity features
Masked autoencoder
Pseudo label
Unsupervised re-ID
Masked autoencoder
Diversity features
Unsupervised re-ID
Pseudo label
Contrastive learning
ISSN:0141-9382, 1872-7387
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Unsupervised person re-identification (Re-ID) methods have made significant progress by exploiting contrastive learning from unlabeled data. However, the previous approaches including cluster-level or instance-level contrast loss, did not fully explore inherent commonality of each identified individual from unlabeled samples, where the divergence of individual cluster and convergence of different clusters leads to a set of noisy pseudo labels which may result in label noise accumulation. To address this issue, we propose an instance-aware diversity feature generation (IDFG) framework, which can form a stable clustering feature space by exhuming diverse counterparts of given exemplars to update memory dictionary of each cluster, so as to reduce the effect of noisy labels. Specifically, we combines instance segmentation and masked auto-encoder to generate foreground-invariant diversity counterparts of given exemplars to reduce inter-class convergence caused by background similarity between different identification instances. Further, we introduce an instance-aware diversity feature mining module, which gradually creates more reliable clusters to generate more robust pseudo labels by exploiting the compactness and independence of clustering to update the memory dictionary. Extensive experiments demonstrate that the proposed IDFG framework achieves impressive performances of 85.6%, 73.7%, and 31.0% mAP on Market1501, DukeMTMC-reID and MSMT17, respectively. •We propose a novel instance-aware diversity feature generation (IDFG) framework, which transcends the limitations of solely relying on hard positive or negative samples for unsupervised person re-identification.•We introduce instance-aware masked auto-encoder that generates foreground-invariant diversity counterparts of given exemplars to alleviate the instance background interference.•We devise an instance-aware diversity feature mining module, which joints diversity-level contrastive loss to exploit the compactness and independence of clustering to update the memory dictionary.•Extensive experiments validate the superiority of our proposed IDFG compared to the star-of-the-art unsupervised ReID methods on Market1501, DukeMTMC-reID and MSMT17 datasets.
ISSN:0141-9382
1872-7387
DOI:10.1016/j.displa.2024.102717