# Real-Time Object Detection Meets DINOv3

arXiv: https://arxiv.org/abs/2509.20787 · HTML: https://mlsanity.com/paper/2509.20787/

## Key facts

- Cited by 27 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 28 references that are in the map.
- First posted on arXiv 25 Sep 2025.

## Builds on (most cited first, 26 of 26)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [You Only Look Once: Unified, Real-Time Object Detection](https://mlsanity.com/paper/1506.02640/) (2015)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://mlsanity.com/paper/2010.04159/) (2020)
- [GLU Variants Improve Transformer](https://mlsanity.com/paper/2002.05202/) (2020)
- [Root Mean Square Layer Normalization](https://mlsanity.com/paper/1910.07467/) (2019)
- [DINOv3](https://mlsanity.com/paper/2508.10104/) (2025)
- [Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regression](https://mlsanity.com/paper/1902.09630/) (2019)
- [YOLOX: Exceeding YOLO Series in 2021](https://mlsanity.com/paper/2107.08430/) (2021)
- [Exploring Plain Vision Transformer Backbones for Object Detection](https://mlsanity.com/paper/2203.16527/) (2022)
- [Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation](https://mlsanity.com/paper/2012.07177/) (2020)
- [DETRs Beat YOLOs on Real-time Object Detection](https://mlsanity.com/paper/2304.08069/) (2023)
- [YOLOv10: Real-Time End-to-End Object Detection](https://mlsanity.com/paper/2405.14458/) (2024)
- [YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information](https://mlsanity.com/paper/2402.13616/) (2024)
- [YOLOv12: Attention-Centric Real-Time Object Detectors](https://mlsanity.com/paper/2502.12524/) (2025)
- [RT-DETRv2: Improved Baseline with Bag-of-Freebies for Real-Time Detection Transformer](https://mlsanity.com/paper/2407.17140/) (2024)
- [D-FINE: Redefine Regression Task in DETRs as Fine-grained Distribution Refinement](https://mlsanity.com/paper/2410.13842/) (2024)
- [DEIM: DETR with Improved Matching for Fast Convergence](https://mlsanity.com/paper/2412.04234/) (2024)
- [YOLOv6 v3.0: A Full-Scale Reloading](https://mlsanity.com/paper/2301.05586/) (2023)
- [Gold-YOLO: Efficient Object Detector via Gather-and-Distribute Mechanism](https://mlsanity.com/paper/2309.11331/) (2023)
- [PP-YOLO: An Effective and Efficient Implementation of Object Detector](https://mlsanity.com/paper/2007.12099/) (2020)
- [PP-PicoDet: A Better Real-Time Object Detector on Mobile Devices](https://mlsanity.com/paper/2111.00902/) (2021)
- [Edge YOLO: Real-Time Intelligent Object Detection System Based on Edge-Cloud Cooperation in Autonomous Vehicles](https://mlsanity.com/paper/2205.14942/) (2022)
- [Beyond Self-Supervision: A Simple Yet Effective Network Distillation Alternative to Improve Backbones](https://mlsanity.com/paper/2103.05959/) (2021)

## Built on by (most cited first, 27 of 27)

- [Ultralytics YOLO26: Unified Real-Time End-to-End Vision Models](https://mlsanity.com/paper/2606.03748/) (2026)
- [OmniRad: A Radiological Foundation Model for Multi-Task Medical Image Analysis](https://mlsanity.com/paper/2602.04547/) (2026)
- [OV-DEIM: Real-time DETR-Style Open-Vocabulary Object Detection with GridSynthetic Augmentation](https://mlsanity.com/paper/2603.07022/) (2026)
- [EdgeCrafter: Compact ViTs for Edge Dense Prediction via Task-Specialized Distillation](https://mlsanity.com/paper/2603.18739/) (2026)
- [Open-Text Aerial Detection: A Unified Framework For Aerial Visual Grounding And Detection](https://mlsanity.com/paper/2602.07827/) (2026)
- [Don't let the information slip away](https://mlsanity.com/paper/2602.22595/) (2026)
- [DINOv3 Visual Representations for Blueberry Perception Toward Robotic Harvesting](https://mlsanity.com/paper/2603.02419/) (2026)
- [RiO-DETR: DETR for Real-time Oriented Object Detection](https://mlsanity.com/paper/2603.09411/) (2026)
- [Caries DETR: Tooth Structure-aware Prior and Lesion-aware Dynamic Loss Refinement for DETR Based Caries Detection](https://mlsanity.com/paper/2604.23718/) (2026)
- [RT-SDGOD: Real-Time Single-Domain Generalized Object Detection](https://mlsanity.com/paper/2606.09367/) (2026)
- [CP-WSP: A Declarative CP-SAT Framework for Configurable Multi-Constraint Workforce Scheduling](https://mlsanity.com/paper/2607.05177/) (2026)
- [DINOv3-Guided Cross Fusion Framework for Semantic-aware CT generation from MRI and CBCT](https://mlsanity.com/paper/2511.12098/) (2025)
- [OpenGround: Planning-based Online Perception for Open-World 3D Visual Grounding](https://mlsanity.com/paper/2512.23020/) (2025)
- [RoLID-11K: A Dashcam Dataset for Small-Object Roadside Litter Detection](https://mlsanity.com/paper/2601.00398/) (2026)
- [DINOv3 Meets YOLO26 for Weed Detection in Vegetable Crops](https://mlsanity.com/paper/2603.00160/) (2026)
- [Open-Vocabulary vs Supervised Learning Methods for Post-Disaster Visual Scene Understanding](https://mlsanity.com/paper/2603.01324/) (2026)
- [Toward Scalable Co-located Practical Learning: Assisting with Computer Vision and Multimodal Analytics](https://mlsanity.com/paper/2603.13679/) (2026)
- [4th Workshop on Maritime Computer Vision (MaCVi): Challenge Overview](https://mlsanity.com/paper/2604.13244/) (2026)
- [Training-Free Tunnel Defect Inspection and Engineering Interpretation via Visual Recalibration and Entity Reconstruction](https://mlsanity.com/paper/2604.27928/) (2026)
- [TinyFormer: Preserving Tiny Objects in YOLO-DETR Hybrid Real-time Detectors](https://mlsanity.com/paper/2605.25046/) (2026)
- [LEVIRDet: A Million-Scale 159-Category Dataset and Foundation Model for Universal Remote Sensing Object Detection](https://mlsanity.com/paper/2606.25312/) (2026)
- [SemCityLoc: Aerial 6DoF Localization Using Semantic 3D City Models](https://mlsanity.com/paper/2606.27444/) (2026)
- [BladeYOLO: Wind Turbine Blade Defect Detection with Limited Annotations and Weak-Saliency Awareness](https://mlsanity.com/paper/2607.28065/) (2026)
- [RegionDet: A Benchmark for Region Detection Beyond Object Instances](https://mlsanity.com/paper/2608.06850/) (2026)
- [GTR: Gated Token Recurrence for Efficient Dense Prediction](https://mlsanity.com/paper/2609.26590/) (2026)
- [S2Planner: Multi-Scale Semantic Planner for End-to-End Autonomous Driving](https://mlsanity.com/paper/2609.29813/) (2026)
- [LiG-DETR: Local-in-Global Reassembly in Latent Space for Aerial Object Detection](https://mlsanity.com/paper/2610.09511/) (2026)

## Abstract (arXiv)

Driven by the simple and effective Dense O2O, DEIM demonstrates faster convergence and enhanced performance. In this work, we extend it with DINOv3 features, resulting in DEIMv2. DEIMv2 spans eight model sizes from X to Atto, covering GPU, edge, and mobile deployment. For the X, L, M, and S variants, we adopt DINOv3-pretrained or distilled backbones and introduce a Spatial Tuning Adapter (STA), which efficiently converts DINOv3's single-scale output into multi-scale features and complements strong semantics with fine-grained details to enhance detection. For ultra-lightweight models (Nano, Pico, Femto, and Atto), we employ HGNetv2 with depth and width pruning to meet strict resource budgets. Together with a simplified decoder and an upgraded Dense O2O, this unified design enables DEIMv2 to achieve a superior performance-cost trade-off across diverse scenarios, establishing new state-of-the-art results. Notably, our largest model, DEIMv2-X, achieves 57.8 AP with only 50.3 million parameters, surpassing prior X-scale models that require over 60 million parameters for just 56.5 AP. On the compact side, DEIMv2-S is the first sub-10 million model (9.71 million) to exceed the 50 AP milestone on COCO, reaching 50.9 AP. Even the ultra-lightweight DEIMv2-Pico, with just 1.5 million parameters, delivers 38.5 AP, matching YOLOv10-Nano (2.3 million) with around 50 percent fewer parameters. Our code and pre-trained models are available at https://github.com/Intellindust-AI-Lab/DEIMv2