# NMS Strikes Back

arXiv: https://arxiv.org/abs/2212.06137 · HTML: https://mlsanity.com/paper/2212.06137/

## Key facts

- Cited by 33 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 39 references that are in the map.
- First posted on arXiv 12 Dec 2022.

## Builds on (most cited first, 30 of 34)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [You Only Look Once: Unified, Real-Time Object Detection](https://mlsanity.com/paper/1506.02640/) (2015)
- [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://mlsanity.com/paper/2010.04159/) (2020)
- [YOLOv3: An Incremental Improvement](https://mlsanity.com/paper/1804.02767/) (2018)
- [FCOS: Fully Convolutional One-Stage Object Detection](https://mlsanity.com/paper/1904.01355/) (2019)
- [YOLO9000: Better, Faster, Stronger](https://mlsanity.com/paper/1612.08242/) (2016)
- [DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection](https://mlsanity.com/paper/2203.03605/) (2022)
- [Cascade R-CNN: Delving into High Quality Object Detection](https://mlsanity.com/paper/1712.00726/) (2017)
- [LVIS: A Dataset for Large Vocabulary Instance Segmentation](https://mlsanity.com/paper/1908.03195/) (2019)
- [Objects as Points](https://mlsanity.com/paper/1904.07850/) (2019)
- [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://mlsanity.com/paper/2107.06278/) (2021)
- [EfficientDet: Scalable and Efficient Object Detection](https://mlsanity.com/paper/1911.09070/) (2019)
- [Focal Loss for Dense Object Detection](https://mlsanity.com/paper/1708.02002/) (2017)
- [Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation](https://mlsanity.com/paper/2012.07177/) (2020)
- [DAB-DETR: Dynamic Anchor Boxes are Better Queries for DETR](https://mlsanity.com/paper/2201.12329/) (2022)
- [Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selection](https://mlsanity.com/paper/1912.02424/) (2019)
- [DN-DETR: Accelerate DETR Training by Introducing Query DeNoising](https://mlsanity.com/paper/2203.01305/) (2022)
- [Hybrid Task Cascade for Instance Segmentation](https://mlsanity.com/paper/1901.07518/) (2019)
- [Conditional DETR for Fast Training Convergence](https://mlsanity.com/paper/2108.06152/) (2021)
- [Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detection](https://mlsanity.com/paper/2006.04388/) (2020)
- [CrowdHuman: A Benchmark for Detecting Human in a Crowd](https://mlsanity.com/paper/1805.00123/) (2018)
- [Scaled-YOLOv4: Scaling Cross Stage Partial Network](https://mlsanity.com/paper/2011.08036/) (2020)
- [Dynamic Head: Unifying Object Detection Heads with Attentions](https://mlsanity.com/paper/2106.08322/) (2021)
- [Rethinking Transformer-based Set Prediction for Object Detection](https://mlsanity.com/paper/2011.10881/) (2020)
- [DETRs with Hybrid Matching](https://mlsanity.com/paper/2207.13080/) (2022)
- [Probabilistic two-stage detection](https://mlsanity.com/paper/2103.07461/) (2021)

## Built on by (most cited first, 30 of 33)

- [Contextual Object Detection with Multimodal Large Language Models](https://mlsanity.com/paper/2305.18279/) (2023)
- [Aligning and Prompting Everything All at Once for Universal Visual Perception](https://mlsanity.com/paper/2312.02153/) (2023)
- [Detection Transformer with Stable Matching](https://mlsanity.com/paper/2304.04742/) (2023)
- [LW-DETR: A Transformer Replacement to YOLO for Real-Time Detection](https://mlsanity.com/paper/2406.03459/) (2024)
- [MS-DETR: Efficient DETR Training with Mixed Supervision](https://mlsanity.com/paper/2401.03989/) (2024)
- [COCONut: Modernizing COCO Segmentation](https://mlsanity.com/paper/2404.08639/) (2024)
- [The 7th AI City Challenge](https://mlsanity.com/paper/2304.07500/) (2023)
- [Enhanced Training of Query-Based Object Detection via Selective Query Recollection](https://mlsanity.com/paper/2212.07593/) (2022)
- [The 8th AI City Challenge](https://mlsanity.com/paper/2404.09432/) (2024)
- [Transformers in Small Object Detection: A Benchmark and Survey of State-of-the-Art](https://mlsanity.com/paper/2309.04902/) (2023)
- [MI-DETR: An Object Detection Model with Multi-time Inquiries Mechanism](https://mlsanity.com/paper/2503.01463/) (2025)
- [detrex: Benchmarking Detection Transformers](https://mlsanity.com/paper/2306.07265/) (2023)
- [Towards Open-Ended Visual Recognition with Large Language Model](https://mlsanity.com/paper/2311.08400/) (2023)
- [World Model-Based End-to-End Scene Generation for Accident Anticipation in Autonomous Driving](https://mlsanity.com/paper/2507.12762/) (2025)
- [Grab What You Need: Rethinking Complex Table Structure Recognition with Flexible Components Deliberation](https://mlsanity.com/paper/2303.09174/) (2023)
- [ASAG: Building Strong One-Decoder-Layer Sparse Detectors via Adaptive Sparse Anchor Generation](https://mlsanity.com/paper/2308.09242/) (2023)
- [Language-conditioned Detection Transformer](https://mlsanity.com/paper/2311.17902/) (2023)
- [SwinTextSpotter v2: Towards Better Synergy for Scene Text Spotting](https://mlsanity.com/paper/2401.07641/) (2024)
- [DETR Doesn't Need Multi-Scale or Locality Design](https://mlsanity.com/paper/2308.01904/) (2023)
- [Evaluating Vision-Language Models for Zero-Shot Detection, Classification, and Association of Motorcycles, Passengers, and Helmets](https://mlsanity.com/paper/2408.02244/) (2024)
- [Adversarial Attention Perturbations for Large Object Detection Transformers](https://mlsanity.com/paper/2508.02987/) (2025)
- [Mcity Data Engine: Iterative Model Improvement Through Open-Vocabulary Data Selection](https://mlsanity.com/paper/2504.21614/) (2025)
- [Enhancing Your Trained DETRs with Box Refinement](https://mlsanity.com/paper/2307.11828/) (2023)
- [SynCL: A Synergistic Training Strategy with Instance-Aware Contrastive Learning for End-to-End Multi-Camera 3D Tracking](https://mlsanity.com/paper/2411.06780/) (2024)
- [Integrating Diverse Assignment Strategies into DETRs](https://mlsanity.com/paper/2601.09247/) (2026)
- [Do We Still Need Non-Maximum Suppression? Accurate Confidence Estimates and Implicit Duplication Modeling with IoU-Aware Calibration](https://mlsanity.com/paper/2309.03110/) (2023)
- [Mr. DETR++: Instructive Multi-Route Training for Detection Transformers with Mixture-of-Experts](https://mlsanity.com/paper/2412.10028/) (2024)
- [MDS-DETR: DETR with Masked Duplicate Suppressor](https://mlsanity.com/paper/2605.23507/) (2026)
- [WoodYOLO: A Novel Object Detector for Wood Species Detection in Microscopic Images](https://mlsanity.com/paper/2411.11738/) (2024)
- [Object Detection Approaches to Identifying Hand Images with High Forensic Values](https://mlsanity.com/paper/2412.16431/) (2024)

## Abstract (arXiv)

Detection Transformer (DETR) directly transforms queries to unique objects by using one-to-one bipartite matching during training and enables end-to-end object detection. Recently, these models have surpassed traditional detectors on COCO with undeniable elegance. However, they differ from traditional detectors in multiple designs, including model architecture and training schedules, and thus the effectiveness of one-to-one matching is not fully understood. In this work, we conduct a strict comparison between the one-to-one Hungarian matching in DETRs and the one-to-many label assignments in traditional detectors with non-maximum supervision (NMS). Surprisingly, we observe one-to-many assignments with NMS consistently outperform standard one-to-one matching under the same setting, with a significant gain of up to 2.5 mAP. Our detector that trains Deformable-DETR with traditional IoU-based label assignment achieved 50.2 COCO mAP within 12 epochs (1x schedule) with ResNet50 backbone, outperforming all existing traditional or transformer-based detectors in this setting. On multiple datasets, schedules, and architectures, we consistently show bipartite matching is unnecessary for performant detection transformers. Furthermore, we attribute the success of detection transformers to their expressive transformer architecture. Code is available at https://github.com/jozhang97/DETA.