mlsanityData to 14 Oct 2026

YOLOv12: Attention-Centric Real-Time Object Detectors

arXiv 2502.12524 ↗First posted 18 Feb 2025cs.CV cs.AIcited by 305

Not on the map. Why?

Abstract (arXiv)

Enhancing the network architecture of the YOLO framework has been crucial for a long time, but has focused on CNN-based improvements despite the proven superiority of attention mechanisms in modeling capabilities. This is because attention-based models cannot match the speed of CNN-based models. This paper proposes an attention-centric YOLO framework, namely YOLOv12, that matches the speed of previous CNN-based ones while harnessing the performance benefits of attention mechanisms. YOLOv12 surpasses all popular real-time object detectors in accuracy with competitive speed. For example, YOLOv12-N achieves 40.6% mAP with an inference latency of 1.64 ms on a T4 GPU, outperforming advanced YOLOv10-N / YOLOv11-N by 2.1%/1.2% mAP with a comparable speed. This advantage extends to other model scales. YOLOv12 also surpasses end-to-end real-time detectors that improve DETR, such as RT-DETR / RT-DETRv2: YOLOv12-S beats RT-DETR-R18 / RT-DETRv2-R18 while running 42% faster, using only 36% of the computation and 45% of the parameters. More comparisons are shown in Figure 1.

Key facts. Cited by 305 papers in the map (data through 14 Oct 2026). Builds on 72 of its 79 references that are in the map. First posted on arXiv 18 Feb 2025.

Builds on

72 of 79 references.

References · 72

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  4. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  5. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  6. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  7. 2020End-to-End Object Detection with Transformerscited by 7,414
  8. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,403
  9. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,183
  10. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,852
  11. 2020Training data-efficient image transformers & distillation through attentioncited by 4,586
  12. 2022FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awarenesscited by 3,673
38 more
  1. 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,422
  2. 2023Qwen Technical Reportcited by 3,344
  3. 2018YOLOv3: An Incremental Improvementcited by 2,659
  4. 2023FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioningcited by 2,598
  5. 2023Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasetscited by 2,581
  6. 2021BEiT: BERT Pre-Training of Image Transformerscited by 2,375
  7. 2020Transformers are RNNs: Fast Autoregressive Transformers with Linear Attentioncited by 2,022
  8. 2021Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutionscited by 1,962
  9. 2020YOLOv4: Optimal Speed and Accuracy of Object Detectioncited by 1,630
  10. 2020Rethinking Attention with Performerscited by 1,601
  11. 2020Linformer: Self-Attention with Linear Complexitycited by 1,474
  12. 2016YOLO9000: Better, Faster, Strongercited by 1,431
  13. 2019Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regressioncited by 1,368
  14. 2024Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Modelcited by 1,056
  15. 2022YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectorscited by 982
  16. 2024VMamba: Visual State Space Modelcited by 969
  17. 2021Going deeper with Image Transformerscited by 764
  18. 2018CCNet: Criss-Cross Attention for Semantic Segmentationcited by 718
  19. 2022EVA: Exploring the Limits of Masked Visual Representation Learning at Scalecited by 691
  20. 2022DAB-DETR: Dynamic Anchor Boxes are Better Queries for DETRcited by 573
  21. 2021Twins: Revisiting the Design of Spatial Attention in Vision Transformerscited by 565
  22. 2018Albumentations: fast and flexible image augmentationscited by 560
  23. 2023DETRs Beat YOLOs on Real-time Object Detectioncited by 533
  24. 2021CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windowscited by 527
  25. 2024YOLOv10: Real-Time End-to-End Object Detectioncited by 515
  26. 2022DN-DETR: Accelerate DETR Training by Introducing Query DeNoisingcited by 488
  27. 2021Nystr"omformer: A Nystr"om-Based Algorithm for Approximating Self-Attentioncited by 471
  28. 2019Distance-IoU Loss: Faster and Better Learning for Bounding Box Regressioncited by 459
  29. 2024YOLOv9: Learning What You Want to Learn Using Programmable Gradient Informationcited by 425
  30. 2021Conditional DETR for Fast Training Convergencecited by 424
  31. 2020Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detectioncited by 409
  32. 2019CSPNet: A New Backbone that can Enhance Learning Capability of CNNcited by 381
  33. 2023EVA-02: A Visual Representation for Neon Genesiscited by 381
  34. 2018Efficient Attention: Attention with Linear Complexitiescited by 368
  35. 2019Axial Attention in Multidimensional Transformerscited by 358
  36. 2021TOOD: Task-aligned One-stage Object Detectioncited by 218
  37. 2021OTA: Optimal Transport Assignment for Object Detectioncited by 178
  38. 2025SANA 1.5: Efficient Scaling of Training-Time and Inference-Time Compute in Linear Diffusion Transformercited by 165

50 most cited of 72 All 72 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 138; 2026: 167.

305 papers.

Sort:

By area:

  1. 2025Cosmos-Drive-Dreams: Scalable Synthetic Driving Data Generation with World Foundation Modelscited by 84
  2. 2025YOLO26: Key Architectural Enhancements and Performance Benchmarking for Real-Time Object Detectioncited by 79
  3. 2025YOLOv13: Real-Time Object Detection with Hypergraph-Enhanced Adaptive Visual Perceptioncited by 58
  4. 2026Ultralytics YOLO26: Unified Real-Time End-to-End Vision Modelscited by 57
  5. 2025Real-Time Object Detection Meets DINOv3cited by 27
  6. 2025Ultralytics YOLO Evolution: An Overview of YOLO27, YOLO26, YOLO11, YOLOv8, and YOLOv5 Object Detectors for Computer Vision and Pattern Recognitioncited by 18
  7. 2025Object Detection with Multimodal Large Vision-Language Models: An In-depth Reviewcited by 14
  8. 2025RF-DETR Object Detection vs YOLOv12 : A Study of Transformer-based and CNN-based Architectures for Single-Class and Multi-Class Greenfruit Detection in Complex Orchard Environments Under Label Ambiguitycited by 12
  9. 2025M4-SAR: A Multi-Resolution, Multi-Polarization, Multi-Scene, Multi-Source Dataset and Benchmark for optical-SAR Object Detectioncited by 10
  10. 2025MultiBanana: A Challenging Benchmark for Multi-Reference Text-to-Image Generationcited by 9
  11. 2026YOLO26: An Analysis of NMS-Free End to End Framework for Real-Time Object Detectioncited by 7
  12. 2025Dome-DETR: DETR with Density-Oriented Feature-Query Manipulation for Efficient Tiny Object Detectioncited by 6
38 more
  1. 2025Beyond conventional vision: RGB-event fusion for robust object detection in dynamic traffic scenarioscited by 6
  2. 2025Adaptive Guidance Semantically Enhanced via Multimodal LLM for Edge-Cloud Object Detectioncited by 6
  3. 2025Neptune-X: Active X-to-Maritime Generation for Universal Maritime Object Detectioncited by 6
  4. 2025Counting Hallucinations in Diffusion Modelscited by 6
  5. 2026MoCha:End-to-End Video Character Replacement without Structural Guidancecited by 6
  6. 2025LLM-Guided Evolution: An Autonomous Model Optimization for Object Detectioncited by 5
  7. 2025Strong Baseline: Multi-UAV Tracking via YOLOv12 with BoT-SORT-ReIDcited by 4
  8. 2025SoftHGNN: Soft Hypergraph Neural Networks for General Visual Recognitioncited by 4
  9. 2025AdvReal: Physical Adversarial Patch Generation Framework for Security Evaluation of Object Detection Systemscited by 4
  10. 2025ProstaTD: Bridging Surgical Triplet from Classification to Fully Supervised Detectioncited by 4
  11. 2025USVTrack: USV-Based 4D Radar-Camera Tracking Dataset for Autonomous Driving in Inland Waterwayscited by 4
  12. 2025Butter: Frequency Consistency and Hierarchical Fusion for Autonomous Driving Object Detectioncited by 4
  13. 2025A Computer Vision Pipeline for Individual-Level Behavior Analysis: Benchmarking on the Edinburgh Pig Datasetcited by 4
  14. 2025YOLO-Master: MOE-Accelerated with Specialized Transformers for Enhanced Real-time Detectioncited by 4
  15. 2026AIVD: Adaptive Edge-Cloud Collaboration for Accurate and Efficient Industrial Visual Detectioncited by 4
  16. 2025Improved YOLOv12 with LLM-Generated Synthetic Data for Enhanced Apple Detection and Benchmarking Against YOLOv11 and YOLOv10cited by 3
  17. 2025YOLO-LLTS: Real-Time Low-Light Traffic Sign Detection via Prior-Guided Enhancement and Multibranch Feature Interactioncited by 3
  18. 2025PixelWeb: The First Web GUI Dataset with Pixel-Wise Labelscited by 3
  19. 2025A Decade of You Only Look Once (YOLO) for Object Detection: A Reviewcited by 3
  20. 2025Mcity Data Engine: Iterative Model Improvement Through Open-Vocabulary Data Selectioncited by 3
  21. 2025YOLOv11-RGBT: Towards a Comprehensive Single-Stage Multispectral Object Detection Frameworkcited by 3
  22. 2025Robust Deep Learning for Myocardial Scar Segmentation in Cardiac MRI with Noisy Labelscited by 3
  23. 2025RoundaboutHD: High-Resolution Real-World Urban Environment Benchmark for Multi-Camera Vehicle Trackingcited by 3
  24. 2025Decentralized Vision-Based Autonomous Aerial Wildlife Monitoringcited by 3
  25. 2025AgriDoctor: A Multimodal Intelligent Assistant for Agriculturecited by 3
  26. 2025Uni-Parser Technical Reportcited by 3
  27. 2026EFSI-DETR: Efficient Frequency-Semantic Integration for Real-Time Small Object Detection in UAV Imagerycited by 3
  28. 2026YOLOE-26: Integrating YOLO26 with YOLOE for Real-Time Open-Vocabulary Instance Segmentationcited by 3
  29. 2026SANEval: Open-Vocabulary Compositional Benchmarks with Failure-mode Diagnosiscited by 3
  30. 2026OmniRad: A Radiological Foundation Model for Multi-Task Medical Image Analysiscited by 3
  31. 2026STVG-R1: Incentivizing Instance-Level Reasoning and Grounding in Videos via Reinforcement Learningcited by 3
  32. 2026WildfireVLM: AI-powered Analysis for Early Wildfire Detection and Risk Assessment Using Satellite Imagerycited by 3
  33. 2026EdgeCrafter: Compact ViTs for Edge Dense Prediction via Task-Specialized Distillationcited by 3
  34. 2026LLM-Guided Agentic Floor Plan Parsing for Accessible Indoor Navigation of Blind and Low-Vision Peoplecited by 3
  35. 2026TRACE: Evidence Grounding-Guided Multi-Video Event Understanding and Claim Generationcited by 3
  36. 2026Unification of Closed-Open Industrial Detection Scenarios: New Large-Scale Benchmarks,Challenges and Baselinescited by 3
  37. 2025An Efficient Aerial Image Detection with Variable Receptive Fieldscited by 2
  38. 2025You Sense Only Once Beneath: Ultra-Light Real-Time Underwater Object Detectioncited by 2

50 most cited of 305 All 305 (JSON)