Transferable Joint Attribute-Identity Deep Learning for Unsupervised Person Re-identification

By Jingya Wang, Xiatian Zhu, Shaogang Gong. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018

606 citations recorded by the source corpus.

Most existing person re-identification (re-id) methods require supervised model learning from a separate large set of pairwise labelled training data for every single camera pair. This significantly limits their scalability and usability in real-world large scale deployments with the need for performing re-id across many camera views. To address this scalability problem, we develop a novel deep learning method for transferring the labelled information of an existing dataset to a new unseen (unlabelled) target domain for person re-id without any supervised learning in the target domain. Specifically, we introduce an Transferable Joint Attribute-Identity Deep Learning (TJ-AIDL) for simultaneously learning an attribute-semantic and identity-discriminative feature representation space transferrable to any new (unseen) target domain for re-id tasks without the need for collecting new labelled training data from the target domain (i.e. unsupervised learning in the target domain). Extensive comparative evaluations validate the superiority of this new TJ-AIDL model for unsupervised person re-id over a wide range of state-of-the-art methods on four challenging benchmarks including VIPeR, PRID, Market-1501, and DukeMTMC-ReID.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/7d8d0b47db6078ad30ee09bd18ea4efe15cd1214

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Yi Yang
2. Luc Van Gool
3. Dacheng Tao
4. Ling Shao
5. Liang Zheng
6. Xiaogang Wang
7. Ming-Hsuan Yang
8. Tao A. Xiang

Citations in the corpus

• Generalizing a Person Retrieval Model Hetero- and Homogeneously (2018, methodology): “In [7], a transferable model is proposed to jointly learn attribute-semantic and identity discriminative feature representation for target domain.”
• Omni-Scale Feature Learning for Person Re-Identification (2019, neutral): “7 TJ-AIDL [56] Market1501 + Duke (U ) 44.”
• Robust Anchor Embedding for Unsupervised Video Person re-IDentification in the Wild (2018, background): “Most of previous methods suffer from scalability issues, and it is hard for them to be applied to the large-scale applications [13, 15,21].”
• RGB-Infrared Cross-Modality Person Re-Identification via Joint Pixel and Feature Alignment (2019, methodology): “Existing methods can be grouped into hand-crafted descriptors [22, 36, 19, 34], metric learning methods [43, 16, 20] and deep learning algorithms [42, 11, 28, 21, 44, 30, 15, 35, 33, 18, 17, 1, 31].”
• Unsupervised Person Re-Identification via Multi-Label Classification (2020, methodology): “In traditional multi-label classification methods, sigmoid and logistic regression loss is a common option [39, 5, 30, 19].”
• Mutual Mean-Teaching: Pseudo Label Refinery for Unsupervised Domain Adaptation on Person Re-identification (2020, neutral): “7 TJ-AIDL (Wang et al., 2018) (CVPR’18) 23.”
• Self-Similarity Grouping: A Simple Unsupervised Cross Domain Adaptation Approach for Person Re-Identification (2018, methodology): “Although unsupervised domain adaptation for person reID has been studied extensively [8, 40, 39, 49], there is still more than 25% and 15% performance drop in mAP and Rank-1 accuracy comparing to their fully supervised counterparts.”
• A Bottom-Up Clustering Approach to Unsupervised Person Re-Identification (2019, methodology): “(Wang et al. 2018) propose to learn an attributesemantic and identity discriminative representation from the source dataset, which is transferable to the target domain.”
• Style Normalization and Restitution for Generalizable Person Re-Identification (2020, methodology): “UDAbased ReID methods can be roughly divided into three categories: style transfer [3, 44, 26], attribute recognition [43, 49, 33], and target-domain pseudo label estimation [5, 37, 58, 40, 52, 50].”
• Cluster Contrast for Unsupervised Person Re-Identification (2021, methodology): “The other is the unsupervised domain adaptation (UDA) person re-ID, which first pre-trains a model on the source labeled dataset, and then fine-tunes the model on the target unlabeled dataset [7, 23, 42, 44, 49, 55, 56].”
• Hierarchical Clustering With Hard-Batch Triplet Loss for Person Re-Identification (2019, methodology): “Finally, it achieves good performance and even surpasses some methods of UDA [4, 6, 26].”
• Rectifying Pseudo Label Learning via Uncertainty Estimation for Domain Adaptive Semantic Segmentation (2020, background): “The main challenge in unsupervised domain adaptation is different data distribution between the source domain and the target domain [Fu et al., 2015, Wang et al., 2018].”
• Invariance Matters: Exemplar Memory for Domain Adaptive Person Re-Identification (2019, methodology): “Recent advanced methods address the UDA problem in person re-ID mostly by reducing the gap between the source and target domains on the image-level [7, 30, 1] or the attribute feature-level [29, 16].”
• Dynamic Dual-Attentive Aggregation Learning for Visible-Infrared Person Re-Identification (2020, background): “Existing works have achieved human-level performance on the widely-used datasets with end-to-end deep learning [1,14,15,17,39,54], either by global [17,64] or part-level feature learning [40,39,67].”
• Deep Learning for Person Re-Identification: A Survey and Outlook (2020, methodology): “4 • † TJ-AIDL [109] requires additional attribute annotation.”