RTMDet: An Empirical Study of Designing Real-Time Object Detectors

by   Chengqi Lyu, et al.
Nanyang Technological University

In this paper, we aim to design an efficient real-time object detector that exceeds the YOLO series and is easily extensible for many object recognition tasks such as instance segmentation and rotated object detection. To obtain a more efficient model architecture, we explore an architecture that has compatible capacities in the backbone and neck, constructed by a basic building block that consists of large-kernel depth-wise convolutions. We further introduce soft labels when calculating matching costs in the dynamic label assignment to improve accuracy. Together with better training techniques, the resulting object detector, named RTMDet, achieves 52.8 FPS on an NVIDIA 3090 GPU, outperforming the current mainstream industrial detectors. RTMDet achieves the best parameter-accuracy trade-off with tiny/small/medium/large/extra-large model sizes for various application scenarios, and obtains new state-of-the-art performance on real-time instance segmentation and rotated object detection. We hope the experimental results can provide new insights into designing versatile real-time object detectors for many object recognition tasks. Code and models are released at https://github.com/open-mmlab/mmdetection/tree/3.x/configs/rtmdet.


page 1

page 2

page 3

page 4


YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors

YOLOv7 surpasses all known object detectors in both speed and accuracy i...

SWA Object Detection

Do you want to improve 1.0 AP for your object detector without any infer...

CBNet: A Novel Composite Backbone Network Architecture for Object Detection

In existing CNN based detectors, the backbone network is a very importan...

Look-into-Object: Self-supervised Structure Modeling for Object Recognition

Most object recognition approaches predominantly focus on learning discr...

The Lottery Ticket Hypothesis for Object Recognition

Recognition tasks, such as object recognition and keypoint estimation, h...

Slim-neck by GSConv: A better design paradigm of detector architectures for autonomous vehicles

Object detection is a difficult downstream task in computer vision. For ...

YOLOv6: A Single-Stage Object Detection Framework for Industrial Applications

For years, the YOLO series has been the de facto industry-level standard...

Please sign up or login with your details

Forgot password? Click here to reset