Improved YOLOv7 Method for Real-Time Obstacle Avoidance in Large Autonomous Vehicles

This paper aims to explore the feasibility of applying the YOLOv7 algorithm to large autonomous vehicles. By implementing convolutional kernels of various sizes in the backbone, the model is made lightweight while maintaining high levels across all mainstream evaluation metrics.

Saved in:
Bibliographic Details
Published in2024 International Conference on Consumer Electronics - Taiwan (ICCE-Taiwan) pp. 467 - 468
Main Authors Chung, Ming-An, Chai, Sung-Yun, Hsieh, Ming-Chun, Lin, Chia-Wei, Hsu, Chia-Chun, Huang, Shang-Jui, Chen, Kai-Xiang, Zhang, Zhi-Xuan, Zhang, Jun-Hao
Format Conference Proceeding
LanguageEnglish
Published IEEE 09.07.2024
Subjects
Accuracy
Autonomous Vehicle
Collision avoidance
Kernel
Measurement
Object detection
Obstacle Avoidance
Real-Time
Real-time systems
Robustness
YOLO
Online AccessGet full text

Cover

More Information
Summary:This paper aims to explore the feasibility of applying the YOLOv7 algorithm to large autonomous vehicles. By implementing convolutional kernels of various sizes in the backbone, the model is made lightweight while maintaining high levels across all mainstream evaluation metrics.
ISSN:2575-8284
DOI:10.1109/ICCE-Taiwan62264.2024.10674444