mlsanityData to 7 Oct 2026

Use Your Head: Improving Long-Tail Video Recognition

arXiv 2304.01143 ↗First posted 3 Apr 2023cs.CVcited by 22

Not on the map. Why?

Abstract (arXiv)

This paper presents an investigation into long-tail video recognition. We demonstrate that, unlike naturally-collected video datasets and existing long-tail image benchmarks, current video benchmarks fall short on multiple long-tailed properties. Most critically, they lack few-shot classes in their tails. In response, we propose new video benchmarks that better assess long-tail recognition, by sampling subsets from two datasets: SSv2 and VideoLT. We then propose a method, Long-Tail Mixed Reconstruction, which reduces overfitting to instances from few-shot classes by reconstructing them as weighted combinations of samples from head classes. LMR then employs label mixing to learn robust decision boundaries. It achieves state-of-the-art average class accuracy on EPIC-KITCHENS and the proposed SSv2-LT and VideoLT-LT. Benchmarks and code at: tobyperrett.github.io/lmr

Key facts. Cited by 22 papers in the map (data through 7 Oct 2026). Builds on 60 of its 79 references that are in the map. First posted on arXiv 3 Apr 2023.

Builds on

60 of 79 references.

References · 60

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  3. 2022Flamingo: a Visual Language Model for Few-Shot Learningcited by 4,337
  4. 2017Quo Vadis, Action Recognition? A New Model and the Kinetics Datasetcited by 3,259
  5. 2016Matching Networks for One Shot Learningcited by 2,246
  6. 2019Class-Balanced Loss Based on Effective Number of Samplescited by 1,239
  7. 2019LVIS: A Dataset for Large Vocabulary Instance Segmentationcited by 1,215
  8. 2017The "something something" video database for learning and evaluating visual common sensecited by 1,092
  9. 2019Learning Imbalanced Datasets with Label-Distribution-Aware Margin Losscited by 987
  10. 2019Decoupling Representation and Classifier for Long-Tailed Recognitioncited by 806
  11. 2019Large-Scale Long-Tailed Recognition in an Open Worldcited by 704
  12. 2020Long-tail learning via logit adjustmentcited by 605
38 more
  1. 2017A systematic study of the class imbalance problem in convolutional neural networkscited by 510
  2. 2018Meta-Learning for Semi-Supervised Few-Shot Classificationcited by 456
  3. 2020Balanced Meta-Softmax for Long-Tailed Visual Recognitioncited by 397
  4. 2016YouTube-8M: A Large-Scale Video Classification Benchmarkcited by 324
  5. 2019Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examplescited by 312
  6. 2020Equalization Loss for Long-Tailed Object Recognitioncited by 285
  7. 2020Long-tailed Recognition by Routing Diverse Distribution-Aware Expertscited by 270
  8. 2021Keeping Your Eye on the Ball: Trajectory Attention in Video Transformerscited by 224
  9. 2021Parametric Contrastive Learningcited by 217
  10. 2021Improving Calibration for Long-Tailed Recognitioncited by 212
  11. 2021Distribution Alignment: A Unified Framework for Long-tail Visual Recognitioncited by 198
  12. 2022MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video Recognitioncited by 189
  13. 2020CrossTransformers: spatially-aware few-shot transfercited by 174
  14. 2020Support-set bottlenecks for video-text representation learningcited by 170
  15. 2020Rethinking Class-Balanced Methods for Long-Tailed Visual Recognition from a Domain Adaptation Perspectivecited by 143
  16. 2022Balanced Contrastive Learning for Long-Tailed Visual Recognitioncited by 143
  17. 2020Feature Space Augmentation for Long-Tailed Datacited by 136
  18. 2020M2m: Imbalanced Classification via Major-to-minor Translationcited by 134
  19. 2021Targeted Supervised Contrastive Learning for Long-Tailed Recognitioncited by 132
  20. 2019Few-Shot Video Classification via Temporal Alignmentcited by 124
  21. 2020Deep Representation Learning on Long-tailed Data: A Learnable Embedding Augmentation Perspectivecited by 120
  22. 2022Long-Tailed Recognition via Weight Balancingcited by 107
  23. 2021The Majority Can Help The Minority: Context-rich Minority Oversampling for Long-tailed Classificationcited by 100
  24. 2020Equalization Loss v2: A New Gradient Balance Approach for Long-tailed Object Detectioncited by 99
  25. 2021MetaSAug: Meta Semantic Augmentation for Long-Tailed Visual Recognitioncited by 99
  26. 2021ResLT: Residual Learning for Long-tailed Recognitioncited by 96
  27. 2021Influence-Balanced Loss for Imbalanced Visual Classificationcited by 91
  28. 2021ACE: Ally Complementary Experts for Solving Long-Tailed Recognition in One-Shotcited by 86
  29. 2021Distributional Robustness Loss for Long-tail Learningcited by 84
  30. 2021Temporal-Relational CrossTransformers for Few-Shot Action Recognitioncited by 83
  31. 2019TARN: Temporal Attentive Relation Network for Few-Shot and Zero-Shot Action Recognitioncited by 71
  32. 2020Posterior Re-calibration for Imbalanced Datasetscited by 53
  33. 2021Spatio-temporal Relation Modeling for Few-shot Action Recognitioncited by 52
  34. 2021Trustworthy Long-Tailed Classificationcited by 51
  35. 2018Learning Compositional Representations for Few-Shot Recognitioncited by 50
  36. 2021Adversarial Robustness under Long-Tailed Distributioncited by 50
  37. 2022Hybrid Relation Guided Set Matching for Few-shot Action Recognitioncited by 49
  38. 2022Equalized Focal Loss for Dense Long-Tailed Object Detectioncited by 46

50 most cited of 60 All 60 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 5; 2024: 6; 2025: 7; 2026: 4.

22 papers.

Sort:

By area:

  1. 2023An Outlook into the Future of Egocentric Visioncited by 84
  2. 2024A Systematic Review on Long-Tailed Learningcited by 27
  3. 2023What can a cook in Italy teach a mechanic in India? Action Recognition Generalisation Over Scenarios and Locationscited by 25
  4. 2023Opening the Vocabulary of Egocentric Actionscited by 23
  5. 2024PanAf20K: A Large Video Dataset for Wild Ape Detection and Behaviour Recognitioncited by 22
  6. 2023Eliciting In-Context Learning in Vision-Language Models for Videos Through Curated Data Distributional Propertiescited by 21
  7. 2024ChimpVLM: Ethogram-Enhanced Chimpanzee Behaviour Recognitioncited by 5
  8. 2025The PanAf-FGBG Dataset: Understanding the Impact of Backgrounds in Wildlife Behaviour Recognitioncited by 5
  9. 2024Long-Tail Temporal Action Segmentation with Group-wise Temporal Logit Adjustmentcited by 3
  10. 2025Classifying Long-tailed and Label-noise Data via Disentangling and Unlearningcited by 2
  11. 2025The Role of Video Generation in Enhancing Data-Limited Action Understandingcited by 2
  12. 2025Long-tailed Species Recognition in the NACTI Wildlife Datasetcited by 2
10 more
  1. 2023Deep Visual-Genetic Biometrics for Taxonomic Classification of Rare Speciescited by 1
  2. 2024Transfer-LMR: Heavy-Tail Driving Behavior Recognition in Diverse Traffic Scenarioscited by 1
  3. 2026MM-TS: Multi-Modal Temperature and Margin Schedules for Contrastive Learning with Long-Tail Datacited by 1
  4. 2024PAT: Pixel-wise Adaptive Training for Long-tailed Segmentationcited by 0
  5. 2025Cost-Sensitive Learning for Long-Tailed Temporal Action Segmentationcited by 0
  6. 2025PI-H2T: Enhancing Long-Tailed Visual Recognition with Permutation-Invariant and Head-to-Tail Feature Fusioncited by 0
  7. 2025Distinguishing Visually Similar Actions: Prompt-Guided Semantic Prototype Modulation for Few-Shot Action Recognitioncited by 0
  8. 2026Gen2Balance: Generative Balancing for Long-Tailed Video Action Recognitioncited by 0
  9. 2026Benchmarking NACTI Species Recognition in Long-Tailed Regimescited by 0
  10. 2026Fine-Grained Action Recognition with Cross-Attentive Latent Sparse Expertscited by 0