Domain Adaptive Person Re-Identification via Coupling Optimization

Liu, Xiaobin; Zhang, Shiliang

doi:10.1145/3394171.3413904

Cited by 32 publications

(11 citation statements)

References 58 publications

(116 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most of these works improve the classical selflearning algorithm on not overfitting the pseudo-label errors, by using teacher-student or ensemble of expert models [13,64,61] while other approaches focus on designing efficient sample selection and outlier detection strategies [54,3]. More robust frameworks are also designed by optimizing losses based on distance distributions [20,27], by leveraging local features [11], intra-inter camera features [53,25], the labeled source samples [8], multiple cluster views [10] or attentionbased model [19], or by mixing pseudo-labels with domaintranslation methods [60,48,71,4], online pseudo-label refinery strategy, temporal ensembling and label propagation [62,66] or meta learning [55]. A recent approach, SpCL [14], proposed self-contrastive learning during the training phase, by leveraging the source and target samples.…”

Section: ) Pseudo-labeling Methodsmentioning

confidence: 99%

Improving Unsupervised Domain Adaptive Re-Identification via Source-Guided Selection of Pseudo-Labeling Hyperparameters

Dubourvieux,

Loesch,

Audigier

et al. 2021

Preprint

View full text Add to dashboard Cite

Unsupervised Domain Adaptation (UDA) for re-identification (re-ID) is a challenging task: to avoid a costly annotation of additional data, it aims at transferring knowledge from a domain with annotated data to a domain of interest with only unlabeled data. Pseudo-labeling approaches have proven to be effective for UDA re-ID. However, the effectiveness of these approaches heavily depends on the choice of some hyperparameters (HP) that affect the generation of pseudo-labels by clustering. The lack of annotation in the domain of interest makes this choice non-trivial. Current approaches simply reuse the same empirical value for all adaptation tasks and regardless of the target data representation that changes through pseudo-labeling training phases. As this simplistic choice may limit their performance, we aim at addressing this issue. We propose new theoretical grounds on HP selection for clustering UDA re-ID as well as method of automatic and cyclic HP tuning for pseudo-labeling UDA clustering: HyPASS. HyPASS consists in incorporating two modules in pseudo-labeling methods: (i) HP selection based on a labeled source validation set and (ii) conditional domain alignment of feature discriminativeness to improve HP selection based on source samples. Experiments on commonly used person re-ID and vehicle re-ID datasets show that our proposed HyPASS consistently improves the best state-of-the-art methods in re-ID compared to the commonly used empirical HP setting.

show abstract

Section: ) Pseudo-labeling Methodsmentioning

confidence: 99%

Improving Unsupervised Domain Adaptive Re-Identification via Source-Guided Selection of Pseudo-Labeling Hyperparameters

Dubourvieux,

Loesch,

Audigier

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Note that our method is very different from previous disentanglement-based UDA methods [17,24,21,35,1,10], which aim to learn domain-invariant features with reduced feature distribution discrepancy between the source and target domains. We intend to achieve "ideal" feature augmentation for effective robust/generalizable feature learning based on disentanglement, where both domaininvariant feature and domain-specific feature are made full use of.…”

Section: Introductionmentioning

confidence: 92%

Disentanglement-based Cross-Domain Feature Augmentation for Effective Unsupervised Domain Adaptive Person Re-identification

Zhang¹,

Lan²,

Zeng³

et al. 2021

Preprint

View full text Add to dashboard Cite

Unsupervised domain adaptive (UDA) person reidentification (ReID) aims to transfer the knowledge from the labeled source domain to the unlabeled target domain for person matching. One challenge is how to generate target domain samples with reliable labels for training. To address this problem, we propose a Disentanglement-based Cross-Domain Feature Augmentation (DCDFA) strategy, where the augmented features characterize well the target and source domain data distributions while inheriting reliable identity labels. Particularly, we disentangle each sample feature into a robust domain-invariant/shared feature and a domain-specific feature, and perform cross-domain feature recomposition to enhance the diversity of samples used in the training, with the constraints of cross-domain ReID loss and domain classification loss. Each recomposed feature, obtained based on the domain-invariant feature (which enables a reliable inheritance of identity) and an enhancement from a domain specific feature (which enables the approximation of real distributions), is thus an "ideal" augmentation. Extensive experimental results demonstrate the effectiveness of our method, which achieves the state-ofthe-art performance.

show abstract

“…Person ReID aims to match a person in non-overlapping camera views. Supervised person ReID has achieved outstanding performance over the years of development [13,22,27,34,39]. Recently, unsupervised person ReID is receiving increased attention [7,20,35,41].…”

Section: Related Work 21 Unsupervised Person Re-identificationmentioning

confidence: 99%

Joint Optimization in Edge-Cloud Continuum for Federated Unsupervised Person Re-identification

Zhuang

Wen

Zhang

2021

Proceedings of the 29th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Person re-identification (ReID) aims to re-identify a person from non-overlapping camera views. Since person ReID data contains sensitive personal information, researchers have adopted federated learning, an emerging distributed training method, to mitigate the privacy leakage risks. However, existing studies rely on data labels that are laborious and time-consuming to obtain. We present FedUReID, a federated unsupervised person ReID system to learn person ReID models without any labels while preserving privacy. FedUReID enables in-situ model training on edges with unlabeled data. A cloud server aggregates models from edges instead of centralizing raw data to preserve data privacy. Moreover, to tackle the problem that edges vary in data volumes and distributions, we personalize training in edges with joint optimization of cloud and edge. Specifically, we propose personalized epoch to reassign computation throughout training, personalized clustering to iteratively predict suitable labels for unlabeled data, and personalized update to adapt the server aggregated model to each edge. Extensive experiments on eight person ReID datasets demonstrate that FedUReID not only achieves higher accuracy but also reduces computation cost by 29%. Our FedUReID system with the joint optimization will shed light on implementing federated learning to more multimedia tasks without data labels. CCS CONCEPTS• Computing methodologies → Distributed algorithms; Object identification; Matching; • Information systems → Top-k retrieval in databases.

show abstract

Domain Adaptive Person Re-Identification via Coupling Optimization

Cited by 32 publications

References 58 publications

Improving Unsupervised Domain Adaptive Re-Identification via Source-Guided Selection of Pseudo-Labeling Hyperparameters

Improving Unsupervised Domain Adaptive Re-Identification via Source-Guided Selection of Pseudo-Labeling Hyperparameters

Disentanglement-based Cross-Domain Feature Augmentation for Effective Unsupervised Domain Adaptive Person Re-identification

Joint Optimization in Edge-Cloud Continuum for Federated Unsupervised Person Re-identification

Contact Info

Product

Resources

About