IDM: An Intermediate Domain Module for Domain Adaptive Person Re-ID

Dai, Yongxing; Li, Jun; Sun, Yifan; Tong, Zekun; Zhang, Chi; Duan, Ling-Yu

doi:10.1109/iccv48922.2021.01165

Cited by 111 publications

(40 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…, N } is a class label, and Y is a random variable of y. The neural network is flexible and often used for encoder (feature extractor) for domain adaptation [16,5,7,8,12,10]; therefore, we use neural network encoder in our proposed method GDAMF. The latent variable of x transformed by the neural network is denoted by z ∈ R d 2 .…”

Section: Proposed Methodsmentioning

confidence: 99%

“…In the context of domain adaptation, several methods have been developed to access one intermediate domain to improve the predictive power on the target dataset [16,5,7,8,12,10]. In these methods, generative models are used such as generative adversarial networks (GAN) [24] and normalizing flow [25] and the datasets from the source and target domains are mixed at arbitrary ratios to obtain the intermediate dataset.…”

Section: Related Work 21 Gradual Domain Adaptationmentioning

confidence: 99%

See 1 more Smart Citation

Cost-effective Framework for Gradual Domain Adaptation with Multifidelity

Sagawa¹,

Hino²

2022

Preprint

View full text Add to dashboard Cite

In domain adaptation, when there is a large distance between the source and target domains, the prediction performance will degrade. Gradual domain adaptation is one of the solutions to such an issue, assuming that we have access to intermediate domains, which shift gradually from the source to target domains. In previous works, it was assumed that the number of samples in the intermediate domains is sufficiently large; hence, self-training was possible without the need for labeled data. If access to an intermediate domain is restricted, self-training will fail. Practically, the cost of samples in intermediate domains will vary, and it is natural to consider that the closer an intermediate domain is to the target domain, the higher the cost of obtaining samples from the intermediate domain is. To solve the trade-off between cost and accuracy, we propose a framework that combines multifidelity and active domain adaptation. The effectiveness of the proposed method is evaluated by experiments with both artificial and real-world datasets. Codes are available at https://github.com/ssgw320/gdamf.

show abstract

Section: Proposed Methodsmentioning

confidence: 99%

Section: Related Work 21 Gradual Domain Adaptationmentioning

confidence: 99%

Cost-effective Framework for Gradual Domain Adaptation with Multifidelity

Sagawa¹,

Hino²

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…As manual annotations are expensive and unavailable in real-world applications, unsupervised person ReID has attracted much more attention. Some researchers use extra labeled images to assist the unsupervised training on unlabeled person ReID by transferring labeled images to the unlabeled domains with GAN-based models [11,32,40,57] or narrowing the distribution gap in feature space [12,24,33]. For example, Liu et al [32] use three GAN models to reduce the discrepancy between different domains in illumination, resolution, and camera-view, respectively.…”

Section: Unsupervised Person Re-identificationmentioning

confidence: 99%

Dual Cluster Contrastive learning for Object Re-Identification

Yao¹,

Xu²

2021

Preprint

View full text Add to dashboard Cite

Recently, cluster contrastive learning has been proven effective for person ReID by computing the contrastive loss between the individual feature and the cluster memory. However, existing methods that use the individual feature to momentum update the cluster memory are not robust to the noisy samples, such as the samples with wrong annotated labels or the pseudo-labels. Unlike the individual-based updating mechanism, the centroid-based updating mechanism that applies the mean feature of each cluster to update the cluster memory is robust against minority noisy samples. Therefore, we formulate the individual-based updating and centroid-based updating mechanisms in a unified cluster contrastive framework, named Dual Cluster Contrastive learning (DCC), which maintains two types of memory banks: individual and centroid cluster memory banks. Significantly, the individual cluster memory is momentum updated based on the individual feature. The centroid cluster memory applies the mean feature of each cluter to update the corresponding cluster memory. Besides the vallina contrastive loss for each memory, a consistency constraint is applied to guarantee the consistency of the output of two memories. Note that DCC can be easily applied for unsupervised or supervised person ReID by using ground-truth labels or pseudo-labels generated with clustering method, respectively. Extensive experiments on two benchmarks under supervised person ReID and unsupervised person ReID demonstrate the superior of the proposed DCC. Code

show abstract

“…Based on the IDM in our previous conference version [12], this paper makes an important extension, i.e., reinforcing IDM into IDM++ by further promoting the alignment from source / target domain to intermediate domains. In IDM, when a source and a target-domain feature are mixed up, their identities are accordingly mixed into a new identity (Section 3.2.1).…”

Section: Introductionmentioning

confidence: 99%

Bridging the Source-to-target Gap for Cross-domain Person Re-Identification with Intermediate Domains

Dai¹,

Li²,

Tong³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Cross-domain person re-identification (re-ID), such as unsupervised domain adaptive re-ID (UDA re-ID), aims to transfer the identity-discriminative knowledge from the source to the target domain. Existing methods commonly consider the source and target domains are isolated from each other, i.e., no intermediate status is modeled between the source and target domains. Directly transferring the knowledge between two isolated domains can be very difficult, especially when the domain gap is large. This paper, from a novel perspective, assumes these two domains are not completely isolated, but can be connected through a series of intermediate domains. Instead of directly aligning the source and target domains against each other, we propose to align the source and target domains against their intermediate domains so as to facilitate a smooth knowledge transfer. To discover and utilize these intermediate domains, this paper proposes an Intermediate Domain Module (IDM) and a Mirrors Generation Module (MGM). IDM has two functions: 1) it generates multiple intermediate domains by mixing the hidden-layer features from source and target domains and 2) it dynamically reduces the domain gap between the source / target domain features and the intermediate domain features. While IDM achieves good domain alignment effect, it introduces a side effect, i.e., the mix-up operation may mix the identities into a new identity and lose the original identities. Accordingly, MGM is introduced to compensate the loss of the original identity by mapping the features into the IDM-generated intermediate domains without changing their original identity. It allows to focus on minimizing domain variations to further promote the alignment between the source / target domain and intermediate domains, which reinforces IDM into IDM++. We extensively evaluate our method under both the UDA and domain generalization (DG) scenarios and observe that IDM++ yields consistent (and usually significant) performance improvement for cross-domain re-ID, achieving new state of the art. For example, on the challenging MSMT17 benchmark, IDM++ surpasses the prior state of the art by a large margin (e.g., up to 9.9% and 7.8% rank-1 accuracy) for UDA and DG scenarios, respectively. Code will be publicly available.

show abstract

IDM: An Intermediate Domain Module for Domain Adaptive Person Re-ID

Cited by 111 publications

References 46 publications

Cost-effective Framework for Gradual Domain Adaptation with Multifidelity

Cost-effective Framework for Gradual Domain Adaptation with Multifidelity

Dual Cluster Contrastive learning for Object Re-Identification

Bridging the Source-to-target Gap for Cross-domain Person Re-Identification with Intermediate Domains

Contact Info

Product

Resources

About