mlsanityData to 7 Oct 2026

Probabilistic two-stage detection

arXiv 2103.07461 ↗First posted 12 Mar 2021cs.CVcited by 130

Not on the map. Why?

Abstract (arXiv)

We develop a probabilistic interpretation of two-stage object detection. We show that this probabilistic interpretation motivates a number of common empirical training practices. It also suggests changes to two-stage detection pipelines. Specifically, the first stage should infer proper object-vs-background likelihoods, which should then inform the overall score of the detector. A standard region proposal network (RPN) cannot infer this likelihood sufficiently well, but many one-stage detectors can. We show how to build a probabilistic two-stage detector from any state-of-the-art one-stage detector. The resulting detectors are faster and more accurate than both their one- and two-stage precursors. Our detector achieves 56.4 mAP on COCO test-dev with single-scale testing, outperforming all published results. Using a lightweight backbone, our detector achieves 49.2 mAP on COCO at 33 fps on a Titan Xp, outperforming the popular YOLOv4 model.

Key facts. Cited by 130 papers in the map (data through 7 Oct 2026). Builds on 54 of its 66 references that are in the map. First posted on arXiv 12 Mar 2021.

Builds on

54 of 66 references.

References · 54

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  4. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  5. 2020End-to-End Object Detection with Transformerscited by 7,384
  6. 2017Mask R-CNNcited by 7,259
  7. 2016Feature Pyramid Networks for Object Detectioncited by 5,021
  8. 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,410
  9. 2015SSD: Single Shot MultiBox Detectorcited by 3,125
  10. 2018YOLOv3: An Incremental Improvementcited by 2,652
  11. 2016Aggregated Residual Transformations for Deep Neural Networkscited by 2,540
  12. 2013Rich feature hierarchies for accurate object detection and semantic segmentationcited by 2,160
38 more
  1. 2020YOLOv4: Optimal Speed and Accuracy of Object Detectioncited by 1,624
  2. 2019FCOS: Fully Convolutional One-Stage Object Detectioncited by 1,573
  3. 2016YOLO9000: Better, Faster, Strongercited by 1,429
  4. 2019MMDetection: Open MMLab Detection Toolbox and Benchmarkcited by 1,379
  5. 2019Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regressioncited by 1,361
  6. 2017Cascade R-CNN: Delving into High Quality Object Detectioncited by 1,275
  7. 2019LVIS: A Dataset for Large Vocabulary Instance Segmentationcited by 1,215
  8. 2019Objects as Pointscited by 1,172
  9. 2019EfficientDet: Scalable and Efficient Object Detectioncited by 1,018
  10. 2016Stacked Hourglass Networks for Human Pose Estimationcited by 913
  11. 2018CornerNet: Detecting Objects as Paired Keypointscited by 763
  12. 2018Deformable ConvNets v2: More Deformable, Better Resultscited by 666
  13. 2019Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selectioncited by 539
  14. 2019Hybrid Task Cascade for Instance Segmentationcited by 479
  15. 2019Res2Net: A New Multi-scale Backbone Architecturecited by 468
  16. 2020ResNeSt: Split-Attention Networkscited by 445
  17. 2017Deep Layer Aggregationcited by 436
  18. 2020Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detectioncited by 408
  19. 2020Rethinking Pre-training and Self-trainingcited by 352
  20. 2020Equalization Loss for Long-Tailed Object Recognitioncited by 285
  21. 2019RepPoints: Point Set Representation for Object Detectioncited by 264
  22. 2020DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolutioncited by 222
  23. 2020Scaled-YOLOv4: Scaling Cross Stage Partial Networkcited by 211
  24. 2020FCOS: A simple and strong anchor-free object detectorcited by 171
  25. 2019Scale-Aware Trident Networks for Object Detectioncited by 159
  26. 2019Bottom-up Object Detection by Grouping Extreme and Center Pointscited by 156
  27. 2020Probabilistic Anchor Assignment with IoU Prediction for Object Detectioncited by 150
  28. 2015Relay Backpropagation for Effective Learning of Deep Convolutional Neural Networkscited by 115
  29. 2020Generalized Focal Loss V2: Learning Reliable Localization Quality Estimation for Dense Object Detectioncited by 104
  30. 2020AutoAssign: Differentiable Label Assignment for Dense Object Detectioncited by 100
  31. 2019FreeAnchor: Learning to Match Anchors for Visual Object Detectioncited by 90
  32. 2019Region Proposal by Guided Anchoringcited by 87
  33. 2020Revisiting the Sibling Head in Object Detectorcited by 82
  34. 2019Cascade RPN: Delving into High-Quality Region Proposal Network with Adaptive Convolutioncited by 57
  35. 2019Soft Anchor-Point Object Detectioncited by 42
  36. 2020CentripetalNet: Pursuing High-quality Keypoint Pairs for Object Detectioncited by 41
  37. 2020RepPoints V2: Verification Meets Regression for Object Detectioncited by 40
  38. 2020BorderDet: Border Feature for Dense Object Detectioncited by 39

50 most cited of 54 All 54 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 13; 2022: 32; 2023: 38; 2024: 35; 2025: 7; 2026: 5.

130 papers.

Sort:

By area:

  1. 2023Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Modelscited by 1,047
  2. 2022Exploring Plain Vision Transformer Backbones for Object Detectioncited by 723
  3. 2022EVA: Exploring the Limits of Masked Visual Representation Learning at Scalecited by 685
  4. 2022Detecting Twenty-thousand Classes using Image-level Supervisioncited by 613
  5. 2021RegionCLIP: Region-based Language-Image Pretrainingcited by 612
  6. 2023EVA-02: A Visual Representation for Neon Genesiscited by 377
  7. 2022DiffusionDet: Diffusion Model for Object Detectioncited by 354
  8. 2022Simple Open-Vocabulary Object Detection with Vision Transformerscited by 303
  9. 2021Focal Self-attention for Local-Global Interactions in Vision Transformerscited by 286
  10. 2022DETRs with Collaborative Hybrid Assignments Trainingcited by 273
  11. 2021Dynamic Head: Unifying Object Detection Heads with Attentionscited by 185
  12. 2021Multimodal Virtual Point 3D Detectioncited by 166
38 more
  1. 2021Instances as Queriescited by 157
  2. 2022Bridging the Gap between Object and Image-level Representations for Open-Vocabulary Detectioncited by 145
  3. 2023CORA: Adapting CLIP for Open-Vocabulary Detection with Region Prompting and Anchor Pre-Matchingcited by 133
  4. 2022GRiT: A Generative Region-to-text Transformer for Object Understandingcited by 119
  5. 2022Global Tracking Transformerscited by 110
  6. 2022Learning Object-Language Alignments for Open-Vocabulary Object Detectioncited by 95
  7. 2021You Only Learn One Representation: Unified Network for Multiple Taskscited by 86
  8. 2021TPH-YOLOv5: Improved YOLOv5 Based on Transformer Prediction Head for Object Detection on Drone-captured Scenarioscited by 83
  9. 2023V3Det: Vast Vocabulary Visual Detection Datasetcited by 81
  10. 2022A Survey on Long-Tailed Visual Recognitioncited by 75
  11. 2022MSMDFusion: Fusing LiDAR and Camera at Multiple Scales with Multi-Depth Seeds for 3D Object Detectioncited by 62
  12. 2023Detecting Everything in the Open World: Towards Universal Object Detectioncited by 62
  13. 2023A Survey on Open-Vocabulary Detection and Segmentation: Past, Present, and Futurecited by 59
  14. 2023Aligning and Prompting Everything All at Once for Universal Visual Perceptioncited by 59
  15. 2023Multi-Modal Classifiers for Open-Vocabulary Object Detectioncited by 53
  16. 2022X-Paste: Revisiting Scalable Copy-Paste for Instance Segmentation using CLIP and StableDiffusioncited by 49
  17. 2023CoDet: Co-Occurrence Guided Region-Word Alignment for Open-Vocabulary Object Detectioncited by 46
  18. 2024LLMs Meet VLMs: Boost Open Vocabulary Object Detection with Fine-grained Descriptorscited by 39
  19. 2023Prompt Pre-Training with Twenty-Thousand Classes for Open-Vocabulary Visual Recognitioncited by 35
  20. 2023MosaicFusion: Diffusion Models as Data Augmenters for Large Vocabulary Instance Segmentationcited by 35
  21. 2022BigDetection: A Large-scale Benchmark for Improved Object Detector Pre-trainingcited by 33
  22. 2022NMS Strikes Backcited by 33
  23. 2023Revisiting Few-Shot Object Detection with Vision-Language Modelscited by 32
  24. 2023SA-BEV: Generating Semantic-Aware Bird's-Eye-View Feature for Multi-view 3D Object Detectioncited by 30
  25. 2024InstaGen: Enhancing Object Detection by Training on Synthetic Datasetcited by 26
  26. 2024Learning Background Prompts to Discover Implicit Knowledge for Open Vocabulary Object Detectioncited by 24
  27. 2023DiffusionEngine: Diffusion Model is Scalable Data Engine for Object Detectioncited by 23
  28. 2023Challenges for Monocular 6D Object Pose Estimation in Roboticscited by 21
  29. 2024Samba: Synchronized Set-of-Sequences Modeling for Multiple Object Trackingcited by 20
  30. 2024LaMI-DETR: Open-Vocabulary Detection with Language Model Instructioncited by 19
  31. 2021Improving Object Detection by Label Assignment Distillationcited by 17
  32. 2022Rethinking Few-Shot Object Detection on a Multi-Domain Benchmarkcited by 17
  33. 2023Three ways to improve feature alignment for open vocabulary detectioncited by 17
  34. 2023LiDAR-Camera Panoptic Segmentation via Geometry-Consistent and Semantic-Aware Alignmentcited by 17
  35. 2022MEVID: Multi-view Extended Videos with Identities for Video Person Re-Identificationcited by 16
  36. 2023OmniLabel: A Challenging Benchmark for Language-Based Object Detectioncited by 16
  37. 2023ScaleDet: A Scalable Multi-Dataset Object Detectorcited by 16
  38. 2024Self-Supervised Multi-Object Tracking with Path Consistencycited by 16

50 most cited of 130 All 130 (JSON)