mlsanityData to 7 Oct 2026

Towards Balanced Learning for Instance Recognition

arXiv 2108.10175 ↗First posted 23 Aug 2021cs.CVcited by 9

Not on the map. Why?

Abstract (arXiv)

Instance recognition is rapidly advanced along with the developments of various deep convolutional neural networks. Compared to the architectures of networks, the training process, which is also crucial to the success of detectors, has received relatively less attention. In this work, we carefully revisit the standard training practice of detectors, and find that the detection performance is often limited by the imbalance during the training process, which generally consists in three levels - sample level, feature level, and objective level. To mitigate the adverse effects caused thereby, we propose Libra R-CNN, a simple yet effective framework towards balanced learning for instance recognition. It integrates IoU-balanced sampling, balanced feature pyramid, and objective re-weighting, respectively for reducing the imbalance at sample, feature, and objective level. Extensive experiments conducted on MS COCO, LVIS and Pascal VOC datasets prove the effectiveness of the overall balanced design.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 54 of its 74 references that are in the map. First posted on arXiv 23 Aug 2021.

Builds on

54 of 74 references.

References · 54

  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. 2016Feature Pyramid Networks for Object Detectioncited by 5,021
  5. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,829
  6. 2015SSD: Single Shot MultiBox Detectorcited by 3,125
  7. 2015Fast R-CNNcited by 2,930
  8. 2017Non-local Neural Networkscited by 2,290
  9. 2013Visualizing and Understanding Convolutional Networkscited by 2,278
  10. 2013Rich feature hierarchies for accurate object detection and semantic segmentationcited by 2,160
  11. 2019FCOS: Fully Convolutional One-Stage Object Detectioncited by 1,573
  12. 2018Group Normalizationcited by 1,514
38 more
  1. 2017Deformable Convolutional Networkscited by 1,481
  2. 2016YOLO9000: Better, Faster, Strongercited by 1,429
  3. 2019MMDetection: Open MMLab Detection Toolbox and Benchmarkcited by 1,379
  4. 2017Cascade R-CNN: Delving into High Quality Object Detectioncited by 1,275
  5. 2019Class-Balanced Loss Based on Effective Number of Samplescited by 1,239
  6. 2019LVIS: A Dataset for Large Vocabulary Instance Segmentationcited by 1,215
  7. 2019Objects as Pointscited by 1,172
  8. 2019Learning Imbalanced Datasets with Label-Distribution-Aware Margin Losscited by 987
  9. 2019Decoupling Representation and Classifier for Long-Tailed Recognitioncited by 806
  10. 2018Path Aggregation Network for Instance Segmentationcited by 796
  11. 2018CornerNet: Detecting Objects as Paired Keypointscited by 763
  12. 2016Multi-View 3D Object Detection Network for Autonomous Drivingcited by 721
  13. 2014Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognitioncited by 719
  14. 2016Training Region-based Object Detectors with Online Hard Example Miningcited by 577
  15. 2019Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selectioncited by 539
  16. 2019Object Detection in 20 Years: A Surveycited by 528
  17. 2019Hybrid Task Cascade for Instance Segmentationcited by 479
  18. 2016R-FCN: Object Detection via Region-based Fully Convolutional Networkscited by 414
  19. 2018Deep Learning for Generic Object Detection: A Surveycited by 392
  20. 2017Relation Networks for Object Detectioncited by 305
  21. 2020Equalization Loss for Long-Tailed Object Recognitioncited by 285
  22. 2019Mask Scoring R-CNNcited by 265
  23. 2019RepPoints: Point Set Representation for Object Detectioncited by 264
  24. 2019Libra R-CNN: Towards Balanced Learning for Object Detectioncited by 255
  25. 2017Flow-Guided Feature Aggregation for Video Object Detectioncited by 177
  26. 2015Cost Sensitive Learning of Deep Feature Representations from Imbalanced Datacited by 167
  27. 2014Multiple Object Tracking: A Literature Reviewcited by 165
  28. 2019Bottom-up Object Detection by Grouping Extreme and Center Pointscited by 156
  29. 2020Overcoming Classifier Imbalance for Long-tail Object Detection with Balanced Group Softmaxcited by 152
  30. 2017Single-Shot Refinement Neural Network for Object Detectioncited by 129
  31. 2017An Analysis of Scale Invariance in Object Detection - SNIPcited by 119
  32. 2018Gradient Harmonized Single-stage Detectorcited by 114
  33. 2017Learning non-maximum suppressioncited by 112
  34. 2018SNIPER: Efficient Multi-Scale Trainingcited by 87
  35. 2019Region Proposal by Guided Anchoringcited by 87
  36. 2016A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detectioncited by 83
  37. 2018M2Det: A Single-Shot Object Detector based on Multi-Level Feature Pyramid Networkcited by 76
  38. 2020Boundary-preserving Mask R-CNNcited by 74

50 most cited of 54 All 54 (JSON)

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 3; 2023: 5.

9 papers.

  1. 2021Context-Aware Transfer Attacks for Object Detectioncited by 18
  2. 2023Adjusting Logit in Gaussian Form for Long-Tailed Visual Recognitioncited by 7
  3. 2023Poses of People in Art: A Data Set for Human Pose Estimation in Digital Art Historycited by 6
  4. 2022Visual Fault Detection of Multi-scale Key Components in Freight Trainscited by 5
  5. 2023Spatial-wise Dynamic Distillation for MLP-like Efficient Visual Fault Detection of Freight Trainscited by 1
  6. 2022A Lightweight NMS-free Framework for Real-time Visual Fault Detection System of Freight Trainscited by 0
  7. 2022Learning High-quality Proposals for Acne Detectioncited by 0
  8. 2023Efficient Visual Fault Detection for Freight Train Braking System via Heterogeneous Self Distillation in the Wildcited by 0
  9. 2023Two-Stage Adaptive Network for Semi-Supervised Cross-Domain Crater Detection under Varying Scenario Distributionscited by 0