mlsanityData to 7 Oct 2026

Omni-Scale Feature Learning for Person Re-Identification

arXiv 1905.00953 ↗First posted 2 May 2019cs.CVcited by 292

Not on the map. Why?

Abstract (arXiv)

As an instance-level recognition problem, person re-identification (ReID) relies on discriminative features, which not only capture different spatial scales but also encapsulate an arbitrary combination of multiple scales. We call features of both homogeneous and heterogeneous scales omni-scale features. In this paper, a novel deep ReID CNN is designed, termed Omni-Scale Network (OSNet), for omni-scale feature learning. This is achieved by designing a residual block composed of multiple convolutional streams, each detecting features at a certain scale. Importantly, a novel unified aggregation gate is introduced to dynamically fuse multi-scale features with input-dependent channel-wise weights. To efficiently learn spatial-channel correlations and avoid overfitting, the building block uses pointwise and depthwise convolutions. By stacking such block layer-by-layer, our OSNet is extremely lightweight and can be trained from scratch on existing ReID benchmarks. Despite its small model size, OSNet achieves state-of-the-art performance on six person ReID datasets, outperforming most large-sized models, often by a clear margin. Code and models are available at: urlhttps://github.com/KaiyangZhou/deep-person-reid.

Key facts. Cited by 292 papers in the map (data through 7 Oct 2026). First posted on arXiv 2 May 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 56; 2022: 58; 2023: 58; 2024: 49; 2025: 35; 2026: 36.

292 papers.

Sort:

By area:

  1. 2021Domain Generalization: A Surveycited by 760
  2. 2021Domain Generalization with MixStylecited by 563
  3. 2021TransReID: Transformer-based Object Re-Identificationcited by 432
  4. 2022BoT-SORT: Robust Associations Multi-Pedestrian Trackingcited by 313
  5. 2022CLIP-ReID: Exploiting Vision-Language Model for Image Re-Identification without Concrete Text Labelscited by 177
  6. 2022Exact Feature Distribution Matching for Arbitrary Style Transfer and Domain Generalizationcited by 122
  7. 2021Counterfactual Attention Learning for Fine-Grained Visual Categorization and Re-identificationcited by 114
  8. 2022Clothes-Changing Person Re-identification with RGB Modality Onlycited by 98
  9. 2022Deep Unlearning via Randomized Conditionally Independent Hessianscited by 83
  10. 2021Cloth-Changing Person Re-identification from A Single Image with Gait Prediction and Regularizationcited by 69
  11. 2021Pose-guided Feature Disentangling for Occluded Person Re-identification Based on Transformercited by 68
  12. 2024State Space Model for New-Generation Network Alternative to Transformers: A Surveycited by 59
38 more
  1. 2022NFormer: Robust Person Re-identification with Neighbor Transformercited by 57
  2. 2022DeepPrivacy2: Towards Realistic Full-Body Anonymizationcited by 51
  3. 2021GiT: Graph Interactive Transformer for Vehicle Re-identificationcited by 49
  4. 2023PLIP: Language-Image Pre-training for Person Representation Learningcited by 49
  5. 2021AAformer: Auto-Aligned Transformer for Person Re-Identificationcited by 48
  6. 2023SparseTrack: Multi-Object Tracking by Performing Scene Decomposition based on Pseudo-Depthcited by 48
  7. 2021Large-Scale Spatio-Temporal Person Re-identification: Algorithms and Benchmarkcited by 47
  8. 2021Style Normalization and Restitution for Domain Generalization and Adaptationcited by 46
  9. 2021Generalizable Representation Learning for Mixture Domain Face Anti-Spoofingcited by 46
  10. 2022GIAOTracker: A comprehensive framework for MCMOT with global information and optimizing strategies in VisDrone 2021cited by 46
  11. 2021On the Unreasonable Effectiveness of Centroids in Image Retrievalcited by 45
  12. 2021Combined Depth Space based Architecture Search For Person Re-identificationcited by 44
  13. 2022Deep Frequency Filtering for Domain Generalizationcited by 41
  14. 2021HAT: Hierarchical Aggregation Transformers for Person Re-identificationcited by 39
  15. 2021Graph Sampling Based Deep Metric Learning for Generalizable Person Re-Identificationcited by 37
  16. 2021AXM-Net: Implicit Cross-Modal Feature Alignment for Person Re-identificationcited by 34
  17. 2024Magic Tokens: Select Diverse Tokens for Multi-modal Object Re-Identificationcited by 34
  18. 2022Surround-view Fisheye Camera Perception for Automated Driving: Overview, Survey and Challengescited by 33
  19. 2021TransMatcher: Deep Image Matching Through Transformers for Generalizable Person Re-identificationcited by 32
  20. 2021DyGLIP: A Dynamic Graph Model with Link Prediction for Accurate Multi-Camera Multiple Object Trackingcited by 31
  21. 2023TOP-ReID: Multi-spectral Object Re-Identification with Token Permutationcited by 31
  22. 2023Iterative Scale-Up ExpansionIoU and Deep Features Association for Multi-Object Tracking in Sportscited by 29
  23. 2024DeMo: Decoupled Feature-Based Mixture of Experts for Multi-Modal Object Re-Identificationcited by 29
  24. 2023SynDrone -- Multi-modal UAV Dataset for Urban Scenarioscited by 27
  25. 2021Dynamic Multi-Person Mesh Recovery From Uncalibrated Multi-View Camerascited by 26
  26. 2021Dense Interaction Learning for Video-based Person Re-identificationcited by 24
  27. 2023ReST: A Reconfigurable Spatial-Temporal Graph Model for Multi-Camera Multi-Object Trackingcited by 24
  28. 2024Implicit Discriminative Knowledge Learning for Visible-Infrared Person Re-Identificationcited by 24
  29. 2021Explainable Person Re-Identification with Attribute-guided Metric Distillationcited by 23
  30. 2021Deep learning-based person re-identification methods: A survey and outlook of recent workscited by 23
  31. 2021Tracking People with 3D Representationscited by 22
  32. 2024Transformer for Object Re-Identification: A Surveycited by 21
  33. 2022The State of Aerial Surveillance: A Surveycited by 20
  34. 2023Adaptive Sparse Pairwise Loss for Object Re-Identificationcited by 20
  35. 2023The 7th AI City Challengecited by 20
  36. 2023Region Generation and Assessment Network for Occluded Person Re-Identificationcited by 20
  37. 2024MambaPro: Multi-Modal Object Re-Identification with Mamba Aggregation and Synergistic Promptcited by 20
  38. 2022Multi-Camera Multiple 3D Object Tracking on the Move for Autonomous Vehiclescited by 19

50 most cited of 292 All 292 (JSON)