mlsanityData to 7 Oct 2026

Confidence Propagation Cluster: Unleash Full Potential of Object Detectors

arXiv 2112.00342 ↗First posted 1 Dec 2021cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

It has been a long history that most object detection methods obtain objects by using the non-maximum suppression (NMS) and its improved versions like Soft-NMS to remove redundant bounding boxes. We challenge those NMS-based methods from three aspects: 1) The bounding box with highest confidence value may not be the true positive having the biggest overlap with the ground-truth box. 2) Not only suppression is required for redundant boxes, but also confidence enhancement is needed for those true positives. 3) Sorting candidate boxes by confidence values is not necessary so that full parallelism is achievable. In this paper, inspired by belief propagation (BP), we propose the Confidence Propagation Cluster (CP-Cluster) to replace NMS-based methods, which is fully parallelizable as well as better in accuracy. In CP-Cluster, we borrow the message passing mechanism from BP to penalize redundant boxes and enhance true positives simultaneously in an iterative way until convergence. We verified the effectiveness of CP-Cluster by applying it to various mainstream detectors such as FasterRCNN, SSD, FCOS, YOLOv3, YOLOv5, Centernet etc. Experiments on MS COCO show that our plug and play method, without retraining detectors, is able to steadily improve average mAP of all those state-of-the-art models with a clear margin from 0.3 to 1.9 respectively when compared with NMS-based methods.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 47 of its 60 references that are in the map. First posted on arXiv 1 Dec 2021.

Builds on

47 of 60 references.

References · 47

  1. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  2. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  3. 2020End-to-End Object Detection with Transformerscited by 7,384
  4. 2017Mask R-CNNcited by 7,259
  5. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,829
  6. 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,410
  7. 2015SSD: Single Shot MultiBox Detectorcited by 3,125
  8. 2015Fast R-CNNcited by 2,930
  9. 2018YOLOv3: An Incremental Improvementcited by 2,652
  10. 2018CBAM: Convolutional Block Attention Modulecited by 2,463
  11. 2013Rich feature hierarchies for accurate object detection and semantic segmentationcited by 2,160
  12. 2020YOLOv4: Optimal Speed and Accuracy of Object Detectioncited by 1,624
35 more
  1. 2019FCOS: Fully Convolutional One-Stage Object Detectioncited by 1,573
  2. 2016YOLO9000: Better, Faster, Strongercited by 1,429
  3. 2019MMDetection: Open MMLab Detection Toolbox and Benchmarkcited by 1,379
  4. 2019Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regressioncited by 1,361
  5. 2017Cascade R-CNN: Delving into High Quality Object Detectioncited by 1,275
  6. 2019LVIS: A Dataset for Large Vocabulary Instance Segmentationcited by 1,215
  7. 2019Objects as Pointscited by 1,172
  8. 2019EfficientDet: Scalable and Efficient Object Detectioncited by 1,018
  9. 2018CornerNet: Detecting Objects as Paired Keypointscited by 763
  10. 2014Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognitioncited by 719
  11. 2017DOTA: A Large-scale Dataset for Object Detection in Aerial Imagescited by 622
  12. 2014Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFscited by 552
  13. 2019Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selectioncited by 539
  14. 2019Distance-IoU Loss: Faster and Better Learning for Bounding Box Regressioncited by 454
  15. 2019YOLACT: Real-time Instance Segmentationcited by 441
  16. 2016R-FCN: Object Detection via Region-based Fully Convolutional Networkscited by 414
  17. 2020Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detectioncited by 408
  18. 2017Soft-NMS -- Improving Object Detection With One Line of Codecited by 380
  19. 2019Libra R-CNN: Towards Balanced Learning for Object Detectioncited by 255
  20. 2020Scaled-YOLOv4: Scaling Cross Stage Partial Networkcited by 211
  21. 2018Acquisition of Localization Confidence for Accurate Object Detectioncited by 174
  22. 2018Bounding Box Regression with Uncertainty for Accurate Object Detectioncited by 143
  23. 2017DSSD : Deconvolutional Single Shot Detectorcited by 130
  24. 2021Probabilistic two-stage detectioncited by 130
  25. 2017Single-Shot Refinement Neural Network for Object Detectioncited by 129
  26. 2021You Only Look One-level Featurecited by 116
  27. 2017Learning non-maximum suppressioncited by 112
  28. 2020AutoAssign: Differentiable Label Assignment for Dense Object Detectioncited by 100
  29. 2020Enhancing Geometric Factors in Model Learning and Inference for Object Detection and Instance Segmentationcited by 98
  30. 2021TPH-YOLOv5: Improved YOLOv5 Based on Transformer Prediction Head for Object Detection on Drone-captured Scenarioscited by 83
  31. 2020End-to-End Object Detection with Fully Convolutional Networkcited by 70
  32. 2019Adaptive NMS: Refining Pedestrian Detection in a Crowdcited by 66
  33. 2019Learning to Cluster Faces on an Affinity Graphcited by 50
  34. 2020What Makes for End-to-End Object Detection?cited by 47
  35. 2020SaccadeNet: A Fast and Accurate Object Detectorcited by 8

Built on by

Not cited yet.