Clustering and Dynamic Sampling Based Unsupervised Domain Adaptation for Person Re-Identification

IEEE Trans. Neural Netw. Learning Syst.

Tiwari

et al. 2024

Self Cite

Recently, unsupervised cross-dataset person reidentification (Re-ID) has attracted more and more attention, which aims to transfer knowledge of a labeled source domain to an unlabeled target domain. There are two common frameworks: one is pixel-alignment of transferring low-level knowledge and the other is feature-alignment of transferring high-level knowledge.In this paper, we propose a novel Recurrent Auto-Encoder (RAE) framework to unify these two kinds of methods and inherit their merits. Specifically, the proposed RAE includes three modules, i.e. a feature-transfer module, a pixel-transfer module and a fusion module. The feature-transfer module utilizes an encoder to map source and target images to a shared feature space. In the space, not only features are identity-discriminative, but also the gap between source and target features is reduced. The pixel-transfer module takes a decoder to reconstruct original images with its features. Here, we hope the images reconstructed from target features are in source-style. Thus, the low-level knowledge can be propagated to the target domain. After transferring both high-and low-level knowledge with the two proposed module above, we design another bilinear pooling layer to fuse both kinds of knowledge. Extensive experiments on Market-1501, DukeMTMC-ReID and MSMT17 datasets show that our method significantly outperforms either pixel-alignment or feature-alignment Re-ID methods, and achieves new state-ofthe-art results.

Section: Related Work a Supervised Re-idmentioning

confidence: 99%

Pixel and Feature Transfer Fusion for Unsupervised Cross-Dataset Person Reidentification

Yang

IEEE Trans. Neural Netw. Learning Syst.

Tiwari

et al. 2024

Self Cite

“…With the popularity of deep learning, some researchers have respectively used k-reciprocal nearest neighbor [131], k-means [27] and self-training clustering target domain [132], [139], [255] to estimate labels of the unlabeled target domain. Zhong et al [133] used an encoding and decoding model to learn the invariant characteristics between the domains.…”

Section: ) Data Enhancementmentioning

confidence: 99%

“…In addition, [135], [136], [255] proposed using auxiliary attributes and identity tags for UDA-based person re-identification, which provided a new idea for follow-up researches. These methods achieved great results but ignored the intra-domain changes in the target domain, that is, the differences between cameras in the target domain.…”

Section: ) Data Enhancementmentioning

confidence: 99%

See 1 more Smart Citation

A Comprehensive Overview of Person Re-Identification Approaches

Zhao

et al. 2020

IEEE Access

Person re-identification, identifying and tracking pedestrians in cross-domain monitoring systems, is an important technology in the computer vision field and of real significance for the construction of smart cities. With the development of deep learning techniques, especially convolutional neural networks, this technology has received more extensive attention and improvement in recent years and a large number of noteworthy achievements have emerged. This paper provides a comprehensive overview of person re-identification approaches to assist researchers in quickly understand this field with preference as well as to provide a more structured framework. By reviewing more than 300 re-identification related papers, the focus of these studies is summarized as information extraction, metric learning, post-processing, efficiency improvement, labeling cost reduction, and data type expansion. This classification is then organized based on different technologies, and on this basis, the pros and cons of each technology are analyzed. Moreover, this overview summarizes the difficulties and challenges of re-identification and discusses the possible research directions for reference. INDEX TERMSComputer vision, convolutional neural networks, distance learning, feature extraction, person re-identification.

“…Due to the lacking of semantic information, the methods often have inferior performance. Domain-transfer-based unsupervised ReID methods [17], [20], [26], [40] transfer knowledge from a labeled source domain to a target unlabeled domain. Yang et al [35] aims to learn discriminative 'individual' patch features to solve the transfer gap on the image-level feature by pre-training on a very large dataset to learn the label knowledge.…”

mentioning

confidence: 99%

Multi-View Label Prediction for Unsupervised Learning Person Re-Identification

Yin

IEEE Signal Process. Lett.

Ding

et al. 2021

Person re-identification (ReID) aims to match pedestrian images across disjoint cameras. Existing supervised ReID methods utilize deep networks and train them with identitylabeled images, which suffer from limited annotations. Recently, clustering-based unsupervised ReID attracts more and more attention. It first clusters unlabeled images and assigns cluster index to the pseudo-identity-labels, then trains a ReID model with the pseudo-identity-labels. However, considering the slight inter-class variations and significant intra-class variations, pseudo-identitylabels learned from clustering algorithms are usually noisy and coarse. To alleviate the problems above, besides clustering pseudoidentity-labels, we propose to learn pseudo-patch-labels, which brings two advantages: (1) Patch naturally alleviates the effect of backgrounds, occlusions, and carryings since they usually occupy small parts in images, thus overcome noisy labels. (2) It is plausible that patches from different pedestrians belong to the same pseudoidentity-label. For example, pedestrians have a high probability of wearing either the same shoes or pants but a low possibility of wearing both. The experiments demonstrate our proposed method achieves the best performance by a large margin on both imageand video-based datasets.