mlsanityData to 7 Oct 2026

OH-Former: Omni-Relational High-Order Transformer for Person Re-Identification

arXiv 2109.11159 ↗First posted 23 Sep 2021cs.CVcited by 10

Not on the map. Why?

Abstract (arXiv)

Transformers have shown preferable performance on many vision tasks. However, for the task of person re-identification (ReID), vanilla transformers leave the rich contexts on high-order feature relations under-exploited and deteriorate local feature details, which are insufficient due to the dramatic variations of pedestrians. In this work, we propose an Omni-Relational High-Order Transformer (OH-Former) to model omni-relational features for ReID. First, to strengthen the capacity of visual representation, instead of obtaining the attention matrix based on pairs of queries and isolated keys at each spatial location, we take a step further to model high-order statistics information for the non-local mechanism. We share the attention weights in the corresponding layer of each order with a prior mixing mechanism to reduce the computation cost. Then, a convolution-based local relation perception module is proposed to extract the local relations and 2D position information. The experimental results of our model are superior promising, which show state-of-the-art performance on Market-1501, DukeMTMC, MSMT17 and Occluded-Duke datasets.

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Builds on 43 of its 53 references that are in the map. First posted on arXiv 23 Sep 2021.

Builds on

43 of 53 references.

References · 43

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020End-to-End Object Detection with Transformerscited by 7,384
  3. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
  4. 2020Training data-efficient image transformers & distillation through attentioncited by 4,560
  5. 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,410
  6. 2021SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformerscited by 2,923
  7. 2017Deformable Convolutional Networkscited by 1,481
  8. 2021Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNetcited by 1,096
  9. 2020Pre-Trained Image Processing Transformercited by 891
  10. 2016Performance Measures and a Data Set for Multi-Target, Multi-Camera Trackingcited by 685
  11. 2021CoAtNet: Marrying Convolution and Attention for All Data Sizescited by 566
  12. 2021ConViT: Improving Vision Transformers with Soft Convolutional Inductive Biasescited by 484
31 more
  1. 2017Person Transfer GAN to Bridge Domain Gap for Person Re-Identificationcited by 479
  2. 2017Beyond Part Models: Person Retrieval with Refined Part Pooling (and a Strong Convolutional Baseline)cited by 472
  3. 2021TransReID: Transformer-based Object Re-Identificationcited by 432
  4. 2019Bag of Tricks and A Strong Baseline for Deep Person Re-identificationcited by 429
  5. 2021Conditional Positional Encodings for Vision Transformerscited by 402
  6. 2021CMT: Convolutional Neural Networks Meet Vision Transformerscited by 310
  7. 2018Learning Discriminative Features with Multiple Granularities for Person Re-Identificationcited by 290
  8. 2015Bilinear CNNs for Fine-grained Visual Recognitioncited by 228
  9. 2021LocalViT: Analyzing Locality in Vision Transformerscited by 228
  10. 2018Harmonious Attention Network for Person Re-Identificationcited by 211
  11. 2021Conformer: Local Features Coupling Global Representations for Visual Recognitioncited by 164
  12. 2021Diverse Part Discovery: Occluded Person Re-identification with Part-Aware Transformercited by 119
  13. 2020Identity-Guided Human Semantic Parsing for Person Re-Identificationcited by 118
  14. 2020High-Order Information Matters: Learning Relation and Topology for Occluded Person Re-Identificationcited by 105
  15. 2018Human Semantic Parsing for Person Re-identificationcited by 100
  16. 2021Contextual Transformer Networks for Visual Recognitioncited by 90
  17. 2017HydraPlus-Net: Attentive Deep Features for Pedestrian Analysiscited by 89
  18. 2018Horizontal Pyramid Matching for Person Re-identificationcited by 70
  19. 2020Pose-guided Visible Part Matching for Occluded Person ReIDcited by 69
  20. 2017Towards Faster Training of Global Covariance Pooling Networks by Iterative Matrix Square Root Normalizationcited by 65
  21. 2019Mixed High-Order Attention Network for Person Re-Identificationcited by 65
  22. 2018Deep Spatial Feature Reconstruction for Partial Person Re-identification: Alignment-Free Approachcited by 58
  23. 2019Foreground-aware Pyramid Reconstruction for Alignment-free Occluded Person Re-identificationcited by 55
  24. 2019Semantics-Aligned Representation Learning for Person Re-identificationcited by 51
  25. 2017Pose Invariant Embedding for Deep Person Re-identificationcited by 47
  26. 2017Deep Representation Learning with Part Loss for Person Re-Identificationcited by 44
  27. 2018Modeling Localness for Self-Attention Networkscited by 43
  28. 2019Second-order Non-local Attention Networks for Person Re-identificationcited by 38
  29. 2019Sharing Attention Weights for Fast Transformercited by 34
  30. 2019Relation Network for Person Re-identificationcited by 29
  31. 2019Multi-Scale Body-Part Mask Guided Attention for Person Re-identificationcited by 7

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 4; 2023: 2; 2024: 3.

10 papers.

  1. 2022DearKD: Data-Efficient Early Knowledge Distillation for Vision Transformerscited by 65
  2. 2022Learning Progressive Modality-shared Transformers for Effective Visible-Infrared Person Re-identificationcited by 40
  3. 2022Lifelong Person Re-Identification via Knowledge Refreshing and Consolidationcited by 26
  4. 2024Transformer for Object Re-Identification: A Surveycited by 21
  5. 2023DC-Former: Diverse and Compact Transformer for Person Re-Identificationcited by 15
  6. 2022D3Former: Debiased Dual Distilled Transformer for Incremental Learningcited by 10
  7. 2024Other Tokens Matter: Exploring Global and Local Features of Vision Transformers for Object Re-Identificationcited by 9
  8. 2023Multi-Stage Spatio-Temporal Aggregation Transformer for Video Person Re-identificationcited by 5
  9. 2021HBReID: Harder Batch for Re-identificationcited by 0
  10. 2024PAFormer: Part Aware Transformer for Person Re-identificationcited by 0