EdgeTrim-YOLO: Improved Trim YOLO Framework Tailored for Deployment on Edge Devices

Jielei Xu, Feng Pan, Xinheng Han, Lingzhi Wang, Yuhe Wang, Weixing Li*

*Corresponding author for this work

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

4 Citations (Scopus)

Abstract

A trim YOLO framework tailored for deployment on edge devices, named EdgeTrim-YOLO, is proposed in this study. Given the limited computing resources of edge devices, traditional YOLO frameworks often fall short of meeting the requirements for real-time performance and model efficacy. To address this issue, we conducted deep optimization and customization of the YOLO framework, introducing GhostConv, DFC Attention, and structural re-parameterization training strategies into the native backbone. These modifications significantly reduced the model's complexity and computational burden while maintaining high detection accuracy on the COCO dataset. Experimental results demonstrate that, compared to the original YOLO framework, the proposed trim YOLO framework achieved an increase in inference speed by 22.4 % on CPU (ARM), 8.2% on GPU, and 19.3% on NPU, respectively, while maintaining comparable detection performance to YOLO v5s. This provides an efficient and feasible solution for real-time object detection applications on edge devices.

Original languageEnglish
Title of host publication2024 4th International Conference on Computer Communication and Artificial Intelligence, CCAI 2024
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages113-118
Number of pages6
ISBN (Electronic)9798350362763
DOIs
Publication statusPublished - 2024
Event4th International Conference on Computer Communication and Artificial Intelligence, CCAI 2024 - Xi'an, China
Duration: 24 May 202426 May 2024

Publication series

Name2024 4th International Conference on Computer Communication and Artificial Intelligence, CCAI 2024

Conference

Conference4th International Conference on Computer Communication and Artificial Intelligence, CCAI 2024
Country/TerritoryChina
CityXi'an
Period24/05/2426/05/24

Keywords

  • edge AI
  • model lightweighting
  • object detection
  • YOLO

Fingerprint

Dive into the research topics of 'EdgeTrim-YOLO: Improved Trim YOLO Framework Tailored for Deployment on Edge Devices'. Together they form a unique fingerprint.

Cite this