EdgeYOLO: An Edge-Real-Time Object Detector

Shihan Liu, Junlin Zha, Jian Sun, Zhuo Li, Gang Wang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

9 Citations (Scopus)

Abstract

An efficient, low-complexity, and anchor-free object detector based on the state-of-the-art YOLO framework is proposed in this paper, which can be implemented in real time on edge computing platforms. An enhanced data augmentation method is developed to effectively suppress overfitting during training, and a hybrid random loss function is designed to improve the detection accuracy of small objects. Inspired by FCOS, a lighter and more efficient decoupled head is proposed, and its inference speed can be improved with little loss of precision. Our baseline model can reach the accuracy of 50.6% AP50:95 and 69.8% AP50 in MS COC02017 dataset, 26.9% AP50:95 and 45.4% AP50 in VisDrone2019-DET dataset, and it meets real-time requirements (FPS230) on edge-computing device Nvidia Jetson AGX Xavier. And as is shown in Fig. 1, lighter models with less parameters designed for edge computing devices with lower computing power also show better performances. Our source code, hyper-parameters and model weights are all available at https://github.com/LSH9832/edgeyolo.

Original languageEnglish
Title of host publication2023 42nd Chinese Control Conference, CCC 2023
PublisherIEEE Computer Society
Pages7507-7512
Number of pages6
ISBN (Electronic)9789887581543
DOIs
Publication statusPublished - 2023
Event42nd Chinese Control Conference, CCC 2023 - Tianjin, China
Duration: 24 Jul 202326 Jul 2023

Publication series

NameChinese Control Conference, CCC
Volume2023-July
ISSN (Print)1934-1768
ISSN (Electronic)2161-2927

Conference

Conference42nd Chinese Control Conference, CCC 2023
Country/TerritoryChina
CityTianjin
Period24/07/2326/07/23

Keywords

  • Anchor-free
  • edge-real-time
  • hybrid random loss
  • object detector

Fingerprint

Dive into the research topics of 'EdgeYOLO: An Edge-Real-Time Object Detector'. Together they form a unique fingerprint.

Cite this