VIU-YOLO: Vehicle-Infrastructure-UAV Cooperative Perception with the Multi-Head Self-Attention Detection Model

Yichi Zhang; Yuekun Hei; Xuting Duan; Wenze Zhang; Chunmian Lin; Jianshan Zhou

doi:10.1109/ICUS66297.2025.11295715

VIU-YOLO: Vehicle-Infrastructure-UAV Cooperative Perception with the Multi-Head Self-Attention Detection Model

Yichi Zhang
, Yuekun Hei
, Xuting Duan^*
, Wenze Zhang
, Chunmian Lin
, Jianshan Zhou

^*Corresponding author for this work

School of Transportation Science and Engineering

Beihang University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Multi-perspective cooperative perception is a promising approach to improving the security and accuracy of unmanned systems in intelligent transportation scenarios. However, existing methods predominantly focus on full-view object cooperative within a closed-loop system, struggling with the simultaneous association of objects from multiple perspectives in complex, dynamic environments. To address this, we proposed a Vehicle-Infrastructure-UAV feature fusion method for multi-perspective cooperative perception, named VIU-YOLO, which can effectively complement multiple viewpoint features for accurate and robust object detection. We validated our method on a publicly available air-ground cooperative perception dataset and the results demonstrate both high detection accuracy and low inference latency.

Original language	English
Title of host publication	Proceedings of 2025 IEEE International Conference on Unmanned Systems, ICUS 2025
Editors	Rong Song
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1629-1634
Number of pages	6
ISBN (Electronic)	9798331526726
DOIs	https://doi.org/10.1109/ICUS66297.2025.11295715
State	Published - 2025
Event	2025 IEEE International Conference on Unmanned Systems, ICUS 2025 - Changzhou, China Duration: 18 Sep 2025 → 19 Sep 2025

Publication series

Name	Proceedings of 2025 IEEE International Conference on Unmanned Systems, ICUS 2025

Conference

Conference	2025 IEEE International Conference on Unmanned Systems, ICUS 2025
Country/Territory	China
City	Changzhou
Period	18/09/25 → 19/09/25

Keywords

Air-ground cooperative perception
intelligent transportation system
multi-agent system
object detection
vehicle-to-everything

Access to Document

10.1109/ICUS66297.2025.11295715

Cite this

Zhang, Y., Hei, Y., Duan, X., Zhang, W., Lin, C., & Zhou, J. (2025). VIU-YOLO: Vehicle-Infrastructure-UAV Cooperative Perception with the Multi-Head Self-Attention Detection Model. In R. Song (Ed.), Proceedings of 2025 IEEE International Conference on Unmanned Systems, ICUS 2025 (pp. 1629-1634). (Proceedings of 2025 IEEE International Conference on Unmanned Systems, ICUS 2025). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICUS66297.2025.11295715

Zhang, Yichi ; Hei, Yuekun ; Duan, Xuting et al. / VIU-YOLO : Vehicle-Infrastructure-UAV Cooperative Perception with the Multi-Head Self-Attention Detection Model. Proceedings of 2025 IEEE International Conference on Unmanned Systems, ICUS 2025. editor / Rong Song. Institute of Electrical and Electronics Engineers Inc., 2025. pp. 1629-1634 (Proceedings of 2025 IEEE International Conference on Unmanned Systems, ICUS 2025).

@inproceedings{7142d469b1bc4207a9b06b6a64fdd5a1,

title = "VIU-YOLO: Vehicle-Infrastructure-UAV Cooperative Perception with the Multi-Head Self-Attention Detection Model",

abstract = "Multi-perspective cooperative perception is a promising approach to improving the security and accuracy of unmanned systems in intelligent transportation scenarios. However, existing methods predominantly focus on full-view object cooperative within a closed-loop system, struggling with the simultaneous association of objects from multiple perspectives in complex, dynamic environments. To address this, we proposed a Vehicle-Infrastructure-UAV feature fusion method for multi-perspective cooperative perception, named VIU-YOLO, which can effectively complement multiple viewpoint features for accurate and robust object detection. We validated our method on a publicly available air-ground cooperative perception dataset and the results demonstrate both high detection accuracy and low inference latency.",

keywords = "Air-ground cooperative perception, intelligent transportation system, multi-agent system, object detection, vehicle-to-everything",

author = "Yichi Zhang and Yuekun Hei and Xuting Duan and Wenze Zhang and Chunmian Lin and Jianshan Zhou",

note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 2025 IEEE International Conference on Unmanned Systems, ICUS 2025 ; Conference date: 18-09-2025 Through 19-09-2025",

year = "2025",

doi = "10.1109/ICUS66297.2025.11295715",

language = "英语",

series = "Proceedings of 2025 IEEE International Conference on Unmanned Systems, ICUS 2025",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1629--1634",

editor = "Rong Song",

booktitle = "Proceedings of 2025 IEEE International Conference on Unmanned Systems, ICUS 2025",

address = "美国",

}

Zhang, Y, Hei, Y, Duan, X, Zhang, W, Lin, C & Zhou, J 2025, VIU-YOLO: Vehicle-Infrastructure-UAV Cooperative Perception with the Multi-Head Self-Attention Detection Model. in R Song (ed.), Proceedings of 2025 IEEE International Conference on Unmanned Systems, ICUS 2025. Proceedings of 2025 IEEE International Conference on Unmanned Systems, ICUS 2025, Institute of Electrical and Electronics Engineers Inc., pp. 1629-1634, 2025 IEEE International Conference on Unmanned Systems, ICUS 2025, Changzhou, China, 18/09/25. https://doi.org/10.1109/ICUS66297.2025.11295715

VIU-YOLO: Vehicle-Infrastructure-UAV Cooperative Perception with the Multi-Head Self-Attention Detection Model. / Zhang, Yichi; Hei, Yuekun; Duan, Xuting et al.
Proceedings of 2025 IEEE International Conference on Unmanned Systems, ICUS 2025. ed. / Rong Song. Institute of Electrical and Electronics Engineers Inc., 2025. p. 1629-1634 (Proceedings of 2025 IEEE International Conference on Unmanned Systems, ICUS 2025).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - VIU-YOLO

T2 - 2025 IEEE International Conference on Unmanned Systems, ICUS 2025

AU - Zhang, Yichi

AU - Hei, Yuekun

AU - Duan, Xuting

AU - Zhang, Wenze

AU - Lin, Chunmian

AU - Zhou, Jianshan

PY - 2025

Y1 - 2025

N2 - Multi-perspective cooperative perception is a promising approach to improving the security and accuracy of unmanned systems in intelligent transportation scenarios. However, existing methods predominantly focus on full-view object cooperative within a closed-loop system, struggling with the simultaneous association of objects from multiple perspectives in complex, dynamic environments. To address this, we proposed a Vehicle-Infrastructure-UAV feature fusion method for multi-perspective cooperative perception, named VIU-YOLO, which can effectively complement multiple viewpoint features for accurate and robust object detection. We validated our method on a publicly available air-ground cooperative perception dataset and the results demonstrate both high detection accuracy and low inference latency.

AB - Multi-perspective cooperative perception is a promising approach to improving the security and accuracy of unmanned systems in intelligent transportation scenarios. However, existing methods predominantly focus on full-view object cooperative within a closed-loop system, struggling with the simultaneous association of objects from multiple perspectives in complex, dynamic environments. To address this, we proposed a Vehicle-Infrastructure-UAV feature fusion method for multi-perspective cooperative perception, named VIU-YOLO, which can effectively complement multiple viewpoint features for accurate and robust object detection. We validated our method on a publicly available air-ground cooperative perception dataset and the results demonstrate both high detection accuracy and low inference latency.

KW - Air-ground cooperative perception

KW - intelligent transportation system

KW - multi-agent system

KW - object detection

KW - vehicle-to-everything

UR - https://www.scopus.com/pages/publications/105031889380

U2 - 10.1109/ICUS66297.2025.11295715

DO - 10.1109/ICUS66297.2025.11295715

M3 - 会议稿件

AN - SCOPUS:105031889380

T3 - Proceedings of 2025 IEEE International Conference on Unmanned Systems, ICUS 2025

SP - 1629

EP - 1634

BT - Proceedings of 2025 IEEE International Conference on Unmanned Systems, ICUS 2025

A2 - Song, Rong

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 18 September 2025 through 19 September 2025

ER -

Zhang Y, Hei Y, Duan X, Zhang W, Lin C , Zhou J. VIU-YOLO: Vehicle-Infrastructure-UAV Cooperative Perception with the Multi-Head Self-Attention Detection Model. In Song R, editor, Proceedings of 2025 IEEE International Conference on Unmanned Systems, ICUS 2025. Institute of Electrical and Electronics Engineers Inc. 2025. p. 1629-1634. (Proceedings of 2025 IEEE International Conference on Unmanned Systems, ICUS 2025). doi: 10.1109/ICUS66297.2025.11295715

VIU-YOLO: Vehicle-Infrastructure-UAV Cooperative Perception with the Multi-Head Self-Attention Detection Model

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this