mlsanityData to 7 Oct 2026

Improving Object Detection by Label Assignment Distillation

arXiv 2108.10520 ↗First posted 24 Aug 2021cs.CVcited by 17

Not on the map. Why?

Abstract (arXiv)

Label assignment in object detection aims to assign targets, foreground or background, to sampled regions in an image. Unlike labeling for image classification, this problem is not well defined due to the object's bounding box. In this paper, we investigate the problem from a perspective of distillation, hence we call Label Assignment Distillation (LAD). Our initial motivation is very simple, we use a teacher network to generate labels for the student. This can be achieved in two ways: either using the teacher's prediction as the direct targets (soft label), or through the hard labels dynamically assigned by the teacher (LAD). Our experiments reveal that: (i) LAD is more effective than soft-label, but they are complementary. (ii) Using LAD, a smaller teacher can also improve a larger student significantly, while soft-label can't. We then introduce Co-learning LAD, in which two networks simultaneously learn from scratch and the role of teacher and student are dynamically interchanged. Using PAA-ResNet50 as a teacher, our LAD techniques can improve detectors PAA-ResNet101 and PAA-ResNeXt101 to 46 rm AP and 47.5rm AP on the COCO test-dev set. With a stronger teacher PAA-SwinB, we improve the students PAA-ResNet50 to 43.7rm AP by only 1x schedule training and standard setting, and PAA-ResNet101 to 47.9rm AP, significantly surpassing the current methods. Our source code and checkpoints are released at https://git.io/JrDZo.

Key facts. Cited by 17 papers in the map (data through 7 Oct 2026). Builds on 46 of its 55 references that are in the map. First posted on arXiv 24 Aug 2021.

Builds on

46 of 55 references.

References · 46

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  3. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  4. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  5. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  6. 2020End-to-End Object Detection with Transformerscited by 7,384
  7. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  8. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
  9. 2016Feature Pyramid Networks for Object Detectioncited by 5,021
  10. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,829
  11. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
  12. 2015SSD: Single Shot MultiBox Detectorcited by 3,125
34 more
  1. 2018YOLOv3: An Incremental Improvementcited by 2,652
  2. 2020YOLOv4: Optimal Speed and Accuracy of Object Detectioncited by 1,624
  3. 2019FCOS: Fully Convolutional One-Stage Object Detectioncited by 1,573
  4. 2019MMDetection: Open MMLab Detection Toolbox and Benchmarkcited by 1,379
  5. 2017Cascade R-CNN: Delving into High Quality Object Detectioncited by 1,275
  6. 2016Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfercited by 1,059
  7. 2018CornerNet: Detecting Objects as Paired Keypointscited by 763
  8. 2018Deformable ConvNets v2: More Deformable, Better Resultscited by 666
  9. 2019Improved Knowledge Distillation via Teacher Assistantcited by 567
  10. 2019Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selectioncited by 539
  11. 2021RepVGG: Making VGG-style ConvNets Great Againcited by 520
  12. 2019Cascade R-CNN: High Quality Object Detection and Instance Segmentationcited by 428
  13. 2020Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detectioncited by 408
  14. 2020Conditional Convolutions for Instance Segmentationcited by 314
  15. 2019CondConv: Conditionally Parameterized Convolutions for Efficient Inferencecited by 273
  16. 2019Distilling Object Detectors with Fine-grained Feature Imitationcited by 188
  17. 2021OTA: Optimal Transport Assignment for Object Detectioncited by 177
  18. 2020Probabilistic Anchor Assignment with IoU Prediction for Object Detectioncited by 150
  19. 2019Feature Selective Anchor-Free Module for Single-Shot Object Detectioncited by 139
  20. 2021Probabilistic two-stage detectioncited by 130
  21. 2021You Only Look One-level Featurecited by 116
  22. 2020TIDE: A General Toolbox for Identifying Object Detection Errorscited by 113
  23. 2021General Instance Distillation for Object Detectioncited by 113
  24. 2021Distilling Object Detectors via Decoupled Featurescited by 105
  25. 2020Generalized Focal Loss V2: Learning Reliable Localization Quality Estimation for Dense Object Detectioncited by 104
  26. 2020Dynamic R-CNN: Towards High Quality Object Detection via Dynamic Trainingcited by 103
  27. 2020AutoAssign: Differentiable Label Assignment for Dense Object Detectioncited by 100
  28. 2019Prime Sample Attention in Object Detectioncited by 55
  29. 2019IoU-aware Single-stage Object Detector for Accurate Localizationcited by 36
  30. 2020Distilling Object Detectors with Task Adaptive Regularizationcited by 36
  31. 2019Learning from Noisy Anchors for One-stage Object Detectioncited by 35
  32. 2019Multiple Anchor Learning for Visual Object Detectioncited by 29
  33. 2021IQDet: Instance-wise Quality Distribution Sampling for Object Detectioncited by 20
  34. 2020MetaDistiller: Network Self-Boosting via Meta-Learned Top-Down Distillationcited by 18

Built on by

Cited by year

Citing papers per year: 2022: 5; 2023: 7; 2024: 3; 2025: 2.

17 papers.

Sort:

  1. 2023CrossKD: Cross-Head Knowledge Distillation for Object Detectioncited by 50
  2. 2022Towards Efficient 3D Object Detection with Knowledge Distillationcited by 29
  3. 2022RaLiBEV: Radar and LiDAR BEV Fusion Learning for Anchor Box Free Object Detection Systemscited by 23
  4. 2022Prediction-Guided Distillation for Dense Object Detectioncited by 14
  5. 2023Bridging Cross-task Protocol Inconsistency for Distillation in Dense Object Detectioncited by 6
  6. 2023Object-centric Cross-modal Feature Distillation for Event-based Object Detectioncited by 6
  7. 2024CTSM: Combining Trait and State Emotions for Empathetic Response Modelcited by 5
  8. 2024ALSS-YOLO: An Adaptive Lightweight Channel Split and Shuffling Network for TIR Wildlife Detection in UAV Imagerycited by 5
  9. 2023Representation Disparity-aware Distillation for 3D Object Detectioncited by 3
  10. 2024LEROjD: Lidar Extended Radar-Only Object Detectioncited by 3
  11. 2022FasterX: Real-Time Object Detection Based on Edge GPUs for UAV Applicationscited by 2
  12. 2023Spatial-wise Dynamic Distillation for MLP-like Efficient Visual Fault Detection of Freight Trainscited by 1
5 more
  1. 2025Architectural Insights into Knowledge Distillation for Object Detection: A Comprehensive Reviewcited by 1
  2. 2022Soft Labels for Rapid Satellite Object Detectioncited by 0
  3. 2023Efficient Visual Fault Detection for Freight Train Braking System via Heterogeneous Self Distillation in the Wildcited by 0
  4. 2023Object Detection for Caries or Pit and Fissure Sealing Requirement in Children's First Permanent Molarscited by 0
  5. 2025From Label Error Detection to Correction: A Modular Framework and Benchmark for Object Detection Datasetscited by 0