mlsanityData to 7 Oct 2026

Distribution Alignment: A Unified Framework for Long-tail Visual Recognition

arXiv 2103.16370 ↗First posted 30 Mar 2021cs.CV cs.AI cs.LGcited by 198

Not on the map. Why?

Abstract (arXiv)

Despite the recent success of deep neural networks, it remains challenging to effectively model the long-tail class distribution in visual recognition tasks. To address this problem, we first investigate the performance bottleneck of the two-stage learning framework via ablative study. Motivated by our discovery, we propose a unified distribution alignment strategy for long-tail visual recognition. Specifically, we develop an adaptive calibration function that enables us to adjust the classification scores for each data point. We then introduce a generalized re-weight method in the two-stage learning to balance the class prior, which provides a flexible and unified solution to diverse scenarios in visual recognition tasks. We validate our method by extensive experiments on four tasks, including image classification, semantic segmentation, object detection, and instance segmentation. Our approach achieves the state-of-the-art results across all four recognition tasks with a simple and unified framework. The code and models will be made publicly available at: https://github.com/Megvii-BaseDetection/DisAlign

Key facts. Cited by 198 papers in the map (data through 7 Oct 2026). Builds on 37 of its 54 references that are in the map. First posted on arXiv 30 Mar 2021.

Builds on

37 of 54 references.

References · 37

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  3. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  4. 2017Mask R-CNNcited by 7,259
  5. 2016Feature Pyramid Networks for Object Detectioncited by 5,021
  6. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,845
  7. 2016DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFscited by 3,239
  8. 2018Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentationcited by 3,210
  9. 2015Fast R-CNNcited by 2,930
  10. 2013Rich feature hierarchies for accurate object detection and semantic segmentationcited by 2,160
  11. 2017Rethinking Atrous Convolution for Semantic Image Segmentationcited by 2,154
  12. 2019Class-Balanced Loss Based on Effective Number of Samplescited by 1,239
25 more
  1. 2019LVIS: A Dataset for Large Vocabulary Instance Segmentationcited by 1,215
  2. 2019Learning Imbalanced Datasets with Label-Distribution-Aware Margin Losscited by 987
  3. 2019Decoupling Representation and Classifier for Long-Tailed Recognitioncited by 806
  4. 2019Large-Scale Long-Tailed Recognition in an Open Worldcited by 704
  5. 2018Learning to Reweight Examples for Robust Deep Learningcited by 670
  6. 2020Long-tail learning via logit adjustmentcited by 605
  7. 2018Exploring the Limits of Weakly Supervised Pretrainingcited by 604
  8. 2017A systematic study of the class imbalance problem in convolutional neural networkscited by 510
  9. 2020ResNeSt: Split-Attention Networkscited by 445
  10. 2019BBN: Bilateral-Branch Network with Cumulative Learning for Long-Tailed Visual Recognitioncited by 403
  11. 2018Context Encoding for Semantic Segmentationcited by 289
  12. 2020Equalization Loss for Long-Tailed Object Recognitioncited by 285
  13. 2020Frustratingly Simple Few-Shot Object Detectioncited by 280
  14. 2020Long-Tailed Classification by Keeping the Good and Removing the Bad Momentum Causal Effectcited by 256
  15. 2018Low-Shot Learning from Imaginary Datacited by 181
  16. 2015Cost Sensitive Learning of Deep Feature Representations from Imbalanced Datacited by 167
  17. 2020Overcoming Classifier Imbalance for Long-tail Object Detection with Balanced Group Softmaxcited by 152
  18. 2020Feature Space Augmentation for Long-Tailed Datacited by 136
  19. 2015Relay Backpropagation for Effective Learning of Deep Convolutional Neural Networkscited by 115
  20. 2020The Devil is in Classification: A Simple Framework for Long-tail Object Detection and Instance Segmentationcited by 108
  21. 2019Striking the Right Balance with Uncertaintycited by 79
  22. 2020Learning to Segment the Tailcited by 54
  23. 2020Solving Long-tailed Recognition with Deep Realistic Taxonomic Classifiercited by 34
  24. 2019Data Augmentation for Object Detection via Progressive and Selective Instance-Switchingcited by 18
  25. 2018Solution for Large-Scale Hierarchical Object Detection Datasets with Incomplete Annotation and Data Imbalancecited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 20; 2022: 45; 2023: 51; 2024: 44; 2025: 26; 2026: 12.

198 papers.

Sort:

By area:

  1. 2022Detecting Twenty-thousand Classes using Image-level Supervisioncited by 613
  2. 2021Deep Long-Tailed Learning: A Surveycited by 416
  3. 2023Towards Open Vocabulary Learning: A Surveycited by 165
  4. 2022Balanced Contrastive Learning for Long-Tailed Visual Recognitioncited by 143
  5. 2021Self-supervised Learning is More Robust to Dataset Imbalancecited by 122
  6. 2022Long-Tailed Recognition via Weight Balancingcited by 107
  7. 2022Retrieval Augmented Classification for Long-Tail Visual Recognitioncited by 106
  8. 2021SGTR: End-to-end Scene Graph Generation with Transformercited by 102
  9. 2021Self-Supervised Aggregation of Diverse Experts for Test-Agnostic Long-Tailed Recognitioncited by 95
  10. 2022A Survey on Long-Tailed Visual Recognitioncited by 75
  11. 2023Long-tailed Visual Recognition via Gaussian Clouded Logit Adjustmentcited by 75
  12. 2022Nested Collaborative Learning for Long-Tailed Visual Recognitioncited by 63
38 more
  1. 2023Long-Tail Learning with Foundation Model: Heavy Fine-Tuning Hurtscited by 58
  2. 2021A Simple Long-Tailed Recognition Baseline via Vision-Language Modelcited by 49
  3. 2022Equalized Focal Loss for Dense Long-Tailed Object Detectioncited by 46
  4. 2023Understanding Imbalanced Semantic Segmentation Through Neural Collapsecited by 46
  5. 2023Global and Local Mixture Consistency Cumulative Learning for Long-tailed Visual Recognitionscited by 43
  6. 2024Probabilistic Contrastive Learning for Long-Tailed Visual Recognitioncited by 42
  7. 2022Towards Long-Tailed 3D Detectioncited by 40
  8. 2021On Model Calibration for Long-Tailed Object Detection and Instance Segmentationcited by 39
  9. 2022Long-Tailed Class Incremental Learningcited by 36
  10. 2023Visually-Prompted Language Model for Fine-Grained Scene Graph Generation in an Open Worldcited by 35
  11. 2023Exploring Vision-Language Models for Imbalanced Learningcited by 35
  12. 2023MosaicFusion: Diffusion Models as Data Augmenters for Large Vocabulary Instance Segmentationcited by 35
  13. 2022Learning Imbalanced Data with Vision Transformerscited by 34
  14. 2021Adaptive Logit Adjustment Loss for Long-Tailed Visual Recognitioncited by 33
  15. 2023ImbSAM: A Closer Look at Sharpness-Aware Minimization in Class-Imbalanced Recognitioncited by 31
  16. 2022Escaping Saddle Points for Effective Generalization on Class-Imbalanced Datacited by 28
  17. 2021Rethinking of Pedestrian Attribute Recognition: A Reliable Evaluation under Zero-Shot Pedestrian Identity Settingcited by 27
  18. 2024A Systematic Review on Long-Tailed Learningcited by 27
  19. 2022Open-Sampling: Exploring Out-of-Distribution data for Re-balancing Long-tailed datasetscited by 25
  20. 2022Constructing Balance from Imbalance for Long-tailed Image Recognitioncited by 25
  21. 2021An EM Framework for Online Incremental Learning of Semantic Segmentationcited by 24
  22. 2023A Unified Perspective for Loss-Oriented Imbalanced Learning via Localizationcited by 24
  23. 2022Learning to Re-weight Examples with Optimal Transport for Imbalanced Classificationcited by 23
  24. 2023Use Your Head: Improving Long-Tail Video Recognitioncited by 22
  25. 2023Long-Tailed Visual Recognition via Self-Heterogeneous Integration with Knowledge Excavationcited by 22
  26. 2022When Noisy Labels Meet Long Tail Dilemmas: A Representation Calibration Methodcited by 21
  27. 2023DiGeo: Discriminative Geometry-Aware Learning for Generalized Few-Shot Object Detectioncited by 21
  28. 2022A Survey of Methods for Addressing Class Imbalance in Deep-Learning Based Natural Language Processingcited by 20
  29. 2021Cross-Domain Empirical Risk Minimization for Unbiased Long-tailed Classificationcited by 19
  30. 2022FEDIC: Federated Learning on Non-IID and Long-Tailed Data via Calibrated Distillationcited by 19
  31. 2023Novel Class Discovery for Long-tailed Recognitioncited by 19
  32. 2022Relieving Long-tailed Instance Segmentation via Pairwise Class Balancecited by 18
  33. 2022Imbalanced Semi-supervised Learning with Bias Adaptive Classifiercited by 18
  34. 2024Delving into the Trajectory Long-tail Distribution for Muti-object Trackingcited by 18
  35. 2024AUCSeg: AUC-oriented Pixel-level Long-tail Semantic Segmentationcited by 18
  36. 2021Margin Calibration for Long-Tailed Visual Recognitioncited by 17
  37. 2022Label-Noise Learning with Intrinsically Long-Tailed Datacited by 17
  38. 2022Towards Calibrated Hyper-Sphere Representation via Distribution Overlap Coefficient for Long-tailed Learningcited by 17

50 most cited of 198 All 198 (JSON)