Generalizable Metric Network for Cross-domain Person Re-identification
Not on the map. Why?
Abstract (arXiv)
Person Re-identification (Re-ID) is a crucial technique for public security and has made significant progress in supervised settings. However, the cross-domain (i.e., domain generalization) scene presents a challenge in Re-ID tasks due to unseen test domains and domain-shift between the training and test sets. To tackle this challenge, most existing methods aim to learn domain-invariant or robust features for all domains. In this paper, we observe that the data-distribution gap between the training and test sets is smaller in the sample-pair space than in the sample-instance space. Based on this observation, we propose a Generalizable Metric Network (GMN) to further explore sample similarity in the sample-pair space. Specifically, we add a Metric Network (M-Net) after the main network and train it on positive and negative sample-pair features, which is then employed during the test stage. Additionally, we introduce the Dropout-based Perturbation (DP) module to enhance the generalization capability of the metric network by enriching the sample-pair diversity. Moreover, we develop a Pair-Identity Center (PIC) loss to enhance the model's discrimination by ensuring that sample-pair features with the same pair-identity are consistent. We validate the effectiveness of our proposed method through a lot of experiments on multiple benchmark datasets and confirm the value of each module in our GMN.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 56 of its 105 references that are in the map. First posted on arXiv 21 Jun 2023.
Builds on
56 of 105 references.
References · 56
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2016Instance Normalization: The Missing Ingredient for Fast Stylizationcited by 1,187
- 2021Generalizing to Unseen Domains: A Survey on Domain Generalizationcited by 762
- 2021Domain Generalization: A Surveycited by 760
- 2018AutoAugment: Learning Augmentation Policies from Datacited by 728
- 2013Domain Generalization via Invariant Feature Representationcited by 622
- 2020Deep Learning for Person Re-identification: A Survey and Outlookcited by 547
- 2017Person Transfer GAN to Bridge Domain Gap for Person Re-Identificationcited by 479
- 2019Domain Generalization by Solving Jigsaw Puzzlescited by 449
- 2018Two at Once: Enhancing Learning and Generalization Capacities via IBN-Netcited by 361
- 2017Re-ranking Person Re-identification with k-reciprocal Encodingcited by 260
- 2019Episodic Training for Domain Generalizationcited by 245
38 more
- 2019Domain Randomization and Pyramid Consistency: Simulation-to-Real Generalization without Accessing Target Domain Datacited by 245
- 2018Harmonious Attention Network for Person Re-Identificationcited by 211
- 2019Joint Discriminative and Generative Learning for Person Re-identificationcited by 211
- 2019Domain-Specific Batch Normalization for Unsupervised Domain Adaptationcited by 197
- 2016Joint Detection and Identification Feature Learning for Person Searchcited by 190
- 2016Person Re-identification: Past, Present and Futurecited by 180
- 2018DLOW: Domain Flow for Adaptation and Generalizationcited by 176
- 2019Learning to Optimize Domain Specific Normalization for Domain Generalizationcited by 160
- 2019A Strong Baseline and Batch Normalization Neck for Deep Person Re-identificationcited by 154
- 2020Style Normalization and Restitution for Generalizable Person Re-identificationcited by 136
- 2020Learning to Generalize Unseen Domains via Memory-based Multi-Source Meta-Learning for Person Re-Identificationcited by 110
- 2019Generalizing to unseen domains via distribution matchingcited by 109
- 2020Unsupervised Pre-training for Person Re-identificationcited by 94
- 2017HydraPlus-Net: Attentive Deep Features for Pedestrian Analysiscited by 89
- 2020AD-Cluster: Augmented Discriminative Clustering for Domain Adaptive Person Re-identificationcited by 81
- 2016A Discriminatively Learned CNN Embedding for Person Re-identificationcited by 78
- 2020Meta Batch-Instance Normalization for Generalizable Person Re-Identificationcited by 77
- 2021Generalizable Person Re-identification with Relevance-aware Mixture of Expertscited by 77
- 2023Towards Unified Text-based Person Retrieval: A Large-scale Multi-Attribute and Language Search Benchmarkcited by 75
- 2019Correlation-aware Adversarial Domain Adaptation and Generalizationcited by 65
- 2021Lifelong Person Re-Identification via Adaptive Knowledge Accumulationcited by 57
- 2019Unsupervised Tracklet Person Re-Identificationcited by 55
- 2019Frustratingly Easy Person Re-Identification: Generalizing Person Re-ID in Practicecited by 48
- 2018Multi-component Image Translation for Deep Domain Generalizationcited by 40
- 2019Learning Disentangled Representation for Robust Person Re-identificationcited by 40
- 2021Feature-based Style Randomization for Domain Generalizationcited by 34
- 2019A Novel Unsupervised Camera-aware Domain Adaptation Framework for Person Re-identificationcited by 32
- 2020Generalizable Model-agnostic Semantic Segmentation via Target-specific Normalizationcited by 20
- 2020Dual Distribution Alignment Network for Generalizable Person Re-Identificationcited by 20
- 2021Mimic Embedding via Adaptive Aggregation: Learning Generalizable Person Re-identificationcited by 19
- 2016Deep Transfer Learning for Person Re-identificationcited by 16
- 2018Where-and-When to Look: Deep Siamese Attention Networks for Video-based Person Re-identificationcited by 14
- 2019Calibrated Domain-Invariant Learning for Highly Generalizable Large Scale Re-Identificationcited by 14
- 2021Deep High-Resolution Representation Learning for Cross-Resolution Person Re-identificationcited by 14
- 2019Cross-Entropy Adversarial View Adaptation for Person Re-identificationcited by 12
- 2019Progressive Cross-camera Soft-label Learning for Semi-supervised Person Re-identificationcited by 12
- 2022A Novel Mix-normalization Method for Generalizable Multi-source Person Re-identificationcited by 11
- 2016PersonNet: Person Re-identification with Deep Convolutional Neural Networkscited by 8
50 most cited of 56 All 56 (JSON)
Built on by
Citing papers per year: 2025: 1; 2026: 1.
2 papers.