mlsanityData to 7 Oct 2026

Soft-NMS -- Improving Object Detection With One Line of Code

arXiv 1704.04503 ↗First posted 14 Apr 2017cs.CVcited by 380

Not on the map. Why?

Abstract (arXiv)

Non-maximum suppression is an integral part of the object detection pipeline. First, it sorts all detection boxes on the basis of their scores. The detection box M with the maximum score is selected and all other detection boxes with a significant overlap (using a pre-defined threshold) with M are suppressed. This process is recursively applied on the remaining boxes. As per the design of the algorithm, if an object lies within the predefined overlap threshold, it leads to a miss. To this end, we propose Soft-NMS, an algorithm which decays the detection scores of all other objects as a continuous function of their overlap with M. Hence, no object is eliminated in this process. Soft-NMS obtains consistent improvements for the coco-style mAP metric on standard datasets like PASCAL VOC 2007 (1.7% for both R-FCN and Faster-RCNN) and MS-COCO (1.3% for R-FCN and 1.1% for Faster-RCNN) by just changing the NMS algorithm without any additional hyper-parameters. Using Deformable-RFCN, Soft-NMS improves state-of-the-art in object detection from 39.8% to 40.9% with a single model. Further, the computational complexity of Soft-NMS is the same as traditional NMS and hence it can be efficiently implemented. Since Soft-NMS does not require any extra training and is simple to implement, it can be easily integrated into any object detection pipeline. Code for Soft-NMS is publicly available on GitHub (http://bit.ly/2nJLNMu).

Key facts. Cited by 380 papers in the map (data through 7 Oct 2026). First posted on arXiv 14 Apr 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 74; 2022: 92; 2023: 69; 2024: 56; 2025: 50; 2026: 39.

380 papers.

Sort:

By area:

  1. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  2. 2021Swin Transformer V2: Scaling Up Capacity and Resolutioncited by 1,215
  3. 2022Exploring Plain Vision Transformer Backbones for Object Detectioncited by 723
  4. 2023ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoderscited by 703
  5. 2022EVA: Exploring the Limits of Masked Visual Representation Learning at Scalecited by 685
  6. 2022Vision Transformer Adapter for Dense Predictionscited by 564
  7. 2021MViTv2: Improved Multiscale Vision Transformers for Classification and Detectioncited by 516
  8. 2023VideoMAE V2: Scaling Video Masked Autoencoders with Dual Maskingcited by 516
  9. 2022Image as a Foreign Language: BEiT Pretraining for All Vision and Vision-Language Taskscited by 510
  10. 2023EVA-02: A Visual Representation for Neon Genesiscited by 377
  11. 2021Pix2seq: A Language Modeling Framework for Object Detectioncited by 345
  12. 2025Perception Encoder: The best visual embeddings are not at the output of the networkcited by 315
38 more
  1. 2021Prompting Visual-Language Models for Efficient Video Understandingcited by 309
  2. 2022ActionFormer: Localizing Moments of Actions with Transformerscited by 308
  3. 2021Focal Self-attention for Local-Global Interactions in Vision Transformerscited by 286
  4. 2022DETRs with Collaborative Hybrid Assignments Trainingcited by 273
  5. 2021Multimodal Motion Prediction with Stacked Transformerscited by 207
  6. 2022AlphaPose: Whole-Body Regional Multi-Person Pose Estimation and Tracking in Real-Timecited by 199
  7. 2021Boundary IoU: Improving Object-Centric Image Segmentation Evaluationcited by 186
  8. 2021Fast Convergence of DETR with Spatially Modulated Co-Attentioncited by 176
  9. 2021FSCE: Few-Shot Object Detection via Contrastive Proposal Encodingcited by 168
  10. 2021Learning Salient Boundary Feature for Anchor-free Temporal Action Localizationcited by 156
  11. 2022Scaling Up Your Kernels to 31x31: Revisiting Large Kernel Design in CNNscited by 155
  12. 2023A Comprehensive Review of YOLO Architectures in Computer Vision: From YOLOv1 to YOLOv8 and YOLO-NAScited by 149
  13. 2023TriDet: Temporal Action Detection with Relative Boundary Modelingcited by 141
  14. 2023ONE-PEACE: Exploring One General Representation Model Toward Unlimited Modalitiescited by 117
  15. 2021Temporal Context Aggregation Network for Temporal Action Proposal Refinementcited by 111
  16. 2021Panoptic SegFormer: Delving Deeper into Panoptic Segmentation with Transformerscited by 109
  17. 2022MonoDETR: Depth-guided Transformer for Monocular 3D Object Detectioncited by 85
  18. 2022AdaMixer: A Fast-Converging Query-Based Object Detectorcited by 78
  19. 2021CBNet: A Composite Backbone Network Architecture for Object Detectioncited by 77
  20. 2021Multimodal Object Detection via Probabilistic Ensemblingcited by 67
  21. 2021GrooMeD-NMS: Grouped Mathematically Differentiable NMS for Monocular 3D Object Detectioncited by 66
  22. 20223D Object Detection from Images for Autonomous Driving: A Surveycited by 66
  23. 2023UnLoc: A Unified Framework for Video Localization Taskscited by 66
  24. 2022Zero-Shot Temporal Action Detection via Vision-Language Promptingcited by 65
  25. 2021Locate and Label: A Two-stage Identifier for Nested Named Entity Recognitioncited by 60
  26. 2022ReAct: Temporal Action Detection with Relational Queriescited by 47
  27. 2022GIAOTracker: A comprehensive framework for MCMOT with global information and optimizing strategies in VisDrone 2021cited by 46
  28. 2023Dense-Localizing Audio-Visual Events in Untrimmed Videos: A Large-Scale Benchmark and Baselinecited by 45
  29. 2023Glitch in the Matrix: A Large Scale Benchmark for Content Driven Audio-Visual Forgery Detection and Localizationcited by 42
  30. 2024End-to-End Autonomous Driving through V2X Cooperationcited by 42
  31. 2021Self-Supervised Learning for Semi-Supervised Temporal Action Proposalcited by 39
  32. 2021Learning Dynamics via Graph Neural Networks for Human Pose Estimation and Trackingcited by 37
  33. 2021DCAN: Improving Temporal Action Detection via Dual Context Aggregationcited by 36
  34. 2022DETRDistill: A Universal Knowledge Distillation Framework for DETR-familiescited by 36
  35. 2024SnAG: Scalable and Accurate Video Groundingcited by 36
  36. 2022Towards Data-Efficient Detection Transformerscited by 34
  37. 2023UMMAFormer: A Universal Multimodal-adaptive Transformer Framework for Temporal Forgery Localizationcited by 33
  38. 2021SOLO: A Simple Framework for Instance Segmentationcited by 32

50 most cited of 380 All 380 (JSON)