mlsanityData to 7 Oct 2026

Learn by Guessing: Multi-Step Pseudo-Label Refinement for Person Re-Identification

arXiv 2101.01215 ↗First posted 4 Jan 2021cs.CV cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Unsupervised Domain Adaptation (UDA) methods for person Re-Identification (Re-ID) rely on target domain samples to model the marginal distribution of the data. To deal with the lack of target domain labels, UDA methods leverage information from labeled source samples and unlabeled target samples. A promising approach relies on the use of unsupervised learning as part of the pipeline, such as clustering methods. The quality of the clusters clearly plays a major role in methods performance, but this point has been overlooked. In this work, we propose a multi-step pseudo-label refinement method to select the best possible clusters and keep improving them so that these clusters become closer to the class divisions without knowledge of the class labels. Our refinement method includes a cluster selection strategy and a camera-based normalization method which reduces the within-domain variations caused by the use of multiple cameras in person Re-ID. This allows our method to reach state-of-the-art UDA results on DukeMTMC-Market1501 (source-target). We surpass state-of-the-art for UDA Re-ID by 3.4% on Market1501-DukeMTMC datasets, which is a more challenging adaptation setup because the target domain (DukeMTMC) has eight distinct cameras. Furthermore, the camera-based normalization method causes a significant reduction in the number of iterations required for training convergence.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 30 of its 49 references that are in the map. First posted on arXiv 4 Jan 2021.

Builds on

30 of 49 references.

References · 30

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2017Random Erasing Data Augmentationcited by 1,235
  3. 2017In Defense of the Triplet Loss for Person Re-Identificationcited by 784
  4. 2017Person Transfer GAN to Bridge Domain Gap for Person Re-Identificationcited by 479
  5. 2019Bag of Tricks and A Strong Baseline for Deep Person Re-identificationcited by 429
  6. 2013DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognitioncited by 422
  7. 2018Two at Once: Enhancing Learning and Generalization Capacities via IBN-Netcited by 361
  8. 2017Unlabeled Samples Generated by GAN Improve the Person Re-identification Baseline in vitrocited by 280
  9. 2017Re-ranking Person Re-identification with k-reciprocal Encodingcited by 260
  10. 2020Mutual Mean-Teaching: Pseudo Label Refinery for Unsupervised Domain Adaptation on Person Re-identificationcited by 214
  11. 2017Image-Image Domain Adaptation with Preserved Self-Similarity and Domain-Dissimilarity for Person Re-identificationcited by 168
  12. 2019A Strong Baseline and Batch Normalization Neck for Deep Person Re-identificationcited by 154
18 more
  1. 2020Style Normalization and Restitution for Generalizable Person Re-identificationcited by 136
  2. 2019Invariance Matters: Exemplar Memory for Domain Adaptive Person Re-identificationcited by 133
  3. 2019Hierarchical Clustering with Hard-batch Triplet Loss for Person Re-identificationcited by 83
  4. 2020Joint Disentangling and Adaptation for Cross-Domain Person Re-Identificationcited by 83
  5. 2020AD-Cluster: Augmented Discriminative Clustering for Domain Adaptive Person Re-identificationcited by 81
  6. 2016A Discriminatively Learned CNN Embedding for Person Re-identificationcited by 78
  7. 2020Unsupervised Person Re-identification via Softened Similarity Learningcited by 75
  8. 2017Unsupervised Person Re-identification: Clustering and Fine-tuningcited by 73
  9. 2018Unsupervised Domain Adaptive Re-Identification: Theory and Practicecited by 69
  10. 2017Camera Style Adaptation for Person Re-identificationcited by 67
  11. 2019Self-training with progressive augmentation for unsupervised cross-domain person re-identificationcited by 54
  12. 2019Frustratingly Easy Person Re-Identification: Generalizing Person Re-ID in Practicecited by 48
  13. 2018Spatial-Temporal Person Re-identificationcited by 45
  14. 2019Cross-Dataset Person Re-Identification via Unsupervised Pose Disentanglement and Adaptationcited by 44
  15. 2017SVDNet for Pedestrian Retrievalcited by 43
  16. 2019A Novel Unsupervised Camera-aware Domain Adaptation Framework for Person Re-identificationcited by 32
  17. 2018EANet: Enhancing Alignment for Cross-Domain Person Re-identificationcited by 16
  18. 2018Adaptation and Re-Identification Network: An Unsupervised Deep Transfer Learning Approach to Person Re-Identificationcited by 9

Built on by

Not cited yet.