DETRs Beat YOLOs on Real-time Object Detection Article Swipe

MAKE IMPACT

Wenyu Lv , Shangliang Xu , Y. Zhao , Guanzhong Wang , Jinman Wei , Cheng Cui , Yuning Du , Qingqing Dang , Yi Liu · YOU? · · 2023 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2304.08069

The YOLO series has become the most popular framework for real-time object detection due to its reasonable trade-off between speed and accuracy. However, we observe that the speed and accuracy of YOLOs are negatively affected by the NMS. Recently, end-to-end Transformer-based detectors (DETRs) have provided an alternative to eliminating NMS. Nevertheless, the high computational cost limits their practicality and hinders them from fully exploiting the advantage of excluding NMS. In this paper, we propose the Real-Time DEtection TRansformer (RT-DETR), the first real-time end-to-end object detector to our best knowledge that addresses the above dilemma. We build RT-DETR in two steps, drawing on the advanced DETR: first we focus on maintaining accuracy while improving speed, followed by maintaining speed while improving accuracy. Specifically, we design an efficient hybrid encoder to expeditiously process multi-scale features by decoupling intra-scale interaction and cross-scale fusion to improve speed. Then, we propose the uncertainty-minimal query selection to provide high-quality initial queries to the decoder, thereby improving accuracy. In addition, RT-DETR supports flexible speed tuning by adjusting the number of decoder layers to adapt to various scenarios without retraining. Our RT-DETR-R50 / R101 achieves 53.1% / 54.3% AP on COCO and 108 / 74 FPS on T4 GPU, outperforming previously advanced YOLOs in both speed and accuracy. We also develop scaled RT-DETRs that outperform the lighter YOLO detectors (S and M models). Furthermore, RT-DETR-R50 outperforms DINO-R50 by 2.2% AP in accuracy and about 21 times in FPS. After pre-training with Objects365, RT-DETR-R50 / R101 achieves 55.3% / 56.2% AP. The project page: https://zhao-yian.github.io/RTDETR.

Type

preprint

Language

en

Landing Page

http://arxiv.org/abs/2304.08069

PDF

https://arxiv.org/pdf/2304.08069

OA Status

green

Cited By

195

Related Works

10

OpenAlex ID

https://openalex.org/W4366400469

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W4366400469

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2304.08069

Digital Object Identifier

Title

DETRs Beat YOLOs on Real-time Object Detection

Work title

Type

preprint

OpenAlex work type

Language

en

Primary language

Publication year

2023

Year of publication

Publication date

2023-04-17

Full publication date if available

Authors

Wenyu Lv, Shangliang Xu, Y. Zhao, Guanzhong Wang, Jinman Wei, Cheng Cui, Yuning Du, Qingqing Dang, Yi Liu

List of authors in order

Landing page

https://arxiv.org/abs/2304.08069

Publisher landing page

PDF URL

https://arxiv.org/pdf/2304.08069

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2304.08069

Direct OA link when available

Concepts

Computer science, Encoder, Detector, Real-time computing, Object detection, Data mining, Speedup, Artificial intelligence, Pattern recognition (psychology), Telecommunications, Operating system

Top concepts (fields/topics) attached by OpenAlex

Cited by

195

Total citation count in OpenAlex

Citations by year (recent)

2025: 86, 2024: 104, 2023: 5

Per-year citation counts (last 5 years)

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload