mlsanityData to 7 Oct 2026

Rethinking Person Re-Identification via Semantic-Based Pretraining

arXiv 2110.05074 ↗First posted 11 Oct 2021cs.CVcited by 6

Not on the map. Why?

Abstract (arXiv)

Pretraining is a dominant paradigm in computer vision. Generally, supervised ImageNet pretraining is commonly used to initialize the backbones of person re-identification (Re-ID) models. However, recent works show a surprising result that CNN-based pretraining on ImageNet has limited impacts on Re-ID system due to the large domain gap between ImageNet and person Re-ID data. To seek an alternative to traditional pretraining, here we investigate semantic-based pretraining as another method to utilize additional textual data against ImageNet pretraining. Specifically, we manually construct a diversified FineGPR-C caption dataset for the first time on person Re-ID events. Based on it, a pure semantic-based pretraining approach named VTBR is proposed to adopt dense captions to learn visual representations with fewer images. We train convolutional neural networks from scratch on the captions of FineGPR-C dataset, and then transfer them to downstream Re-ID tasks. Comprehensive experiments conducted on benchmark datasets show that our VTBR can achieve competitive performance compared with ImageNet pretraining - despite using up to 1.4x fewer images, revealing its potential in Re-ID pretraining.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 40 of its 59 references that are in the map. First posted on arXiv 11 Oct 2021.

Builds on

40 of 59 references.

References · 40

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  4. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  5. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  6. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,845
  7. 2018Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labelscited by 994
  8. 2017In Defense of the Triplet Loss for Person Re-Identificationcited by 784
  9. 2016Performance Measures and a Data Set for Multi-Target, Multi-Camera Trackingcited by 685
  10. 2017Person Transfer GAN to Bridge Domain Gap for Person Re-Identificationcited by 479
  11. 2017Beyond Part Models: Person Retrieval with Refined Part Pooling (and a Strong Convolutional Baseline)cited by 472
  12. 2019Bag of Tricks and A Strong Baseline for Deep Person Re-identificationcited by 429
28 more
  1. 2020Circle Loss: A Unified Perspective of Pair Similarity Optimizationcited by 402
  2. 2018Rethinking ImageNet Pre-trainingcited by 376
  3. 2018Two at Once: Enhancing Learning and Generalization Capacities via IBN-Netcited by 361
  4. 2020VirTex: Learning Visual Representations from Textual Annotationscited by 338
  5. 2019Lookahead Optimizer: k steps forward, 1 step backcited by 333
  6. 2017Unlabeled Samples Generated by GAN Improve the Person Re-identification Baseline in vitrocited by 280
  7. 2017Re-ranking Person Re-identification with k-reciprocal Encodingcited by 260
  8. 2019Joint Discriminative and Generative Learning for Person Re-identificationcited by 211
  9. 2020Eigen-CAM: Class Activation Map using Principal Componentscited by 188
  10. 2017Image-Image Domain Adaptation with Preserved Self-Similarity and Domain-Dissimilarity for Person Re-identificationcited by 168
  11. 2017Improving Person Re-identification by Attribute and Identity Learningcited by 95
  12. 2020Unsupervised Pre-training for Person Re-identificationcited by 94
  13. 2019Hierarchical Clustering with Hard-batch Triplet Loss for Person Re-identificationcited by 83
  14. 2018FD-GAN: Pose-guided Feature Distilling GAN for Robust Person Re-identificationcited by 70
  15. 2017Pose-Normalized Image Generation for Person Re-identificationcited by 69
  16. 2019Mixed High-Order Attention Network for Person Re-Identificationcited by 65
  17. 2017SVDNet for Pedestrian Retrievalcited by 43
  18. 2019Semantic-Aware Scene Recognitioncited by 43
  19. 2018Re-Identification with Consistent Attentive Siamese Networkscited by 40
  20. 2016Gated Siamese Convolutional Neural Network Architecture for Human Re-Identificationcited by 34
  21. 2021Unsupervised Multi-Source Domain Adaptation for Person Re-Identificationcited by 30
  22. 2019Relation Network for Person Re-identificationcited by 29
  23. 2018Resource Aware Person Re-identification across Multiple Resolutionscited by 21
  24. 2021Less is More: Learning from Synthetic Data with Fine-grained Attributes for Person Re-Identificationcited by 19
  25. 2017Margin Sample Mining Loss: A Deep Learning Based Method for Person Re-identificationcited by 12
  26. 2021Learning from Self-Discrepancy via Multiple Co-teaching for Cross-Domain Person Re-Identificationcited by 11
  27. 2021ASMR: Learning Attribute-Based Person Search with Adaptive Semantic Margin Regularizercited by 11
  28. 2020Taking A Closer Look at Synthesis: Fine-grained Attribute Analysis for Person Re-Identificationcited by 6

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 1; 2024: 2; 2025: 2.

6 papers.

  1. 2024VM-UNet: Vision Mamba UNet for Medical Image Segmentationcited by 286
  2. 2023Learning Multi-axis Representation in Frequency Domain for Medical Image Segmentationcited by 3
  3. 2022Deep Multimodal Fusion for Generalizable Person Re-identificationcited by 2
  4. 2024Learning Collaborative Knowledge with Multimodal Representation for Polyp Re-Identificationcited by 1
  5. 2025Hand Held Multi-Object Tracking Dataset in American Footballcited by 0
  6. 2025GPF-Net: Gated Progressive Fusion Learning for Polyp Re-Identificationcited by 0