A Lightweight Vehicle Detection Model for Complex Traffic Scenarios

View at publisher Summarise with VIRAT
Authors: Feifan Liu, Yuhu Shi
Cited by: 0
References: 12

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

With the development of autonomous driving, the detection of pedestrians and vehicles in complex scenes has been challenged by the problems of misdetection, omission and high model complexity. In order to overcome these problems, a novel lightweight target detection algorithm PCA-YOLO is proposed based on YOLO11 in this study. Firstly, partial convolution is integrated into the backbone network, and the C3k2 module is improved into the P_C3k2 structure, which can efficiently extract spatial features while reducing the number of model parameters and computational overhead. Secondly, a lightweight upsampling operator are added to the neck structure to better capture target information; meanwhile, the adaptive spatial feature fusion module is embedded into the object detector head, which effectively filters out conflicting information. Finally, experimental outcomes on the KITTI dataset show that the model PCA-YOLO can attain higher detection accuracy when compared against YOLO11n and other baseline algorithms, with varying degrees of model size, parameter count, and complexity reduction.

Fields of Study & Subjects

Advanced Neural Network Applications Autonomous Vehicle Technology and Safety Domain Adaptation and Few-Shot Learning

Publication details

Journal / source
International Conference on Communication Networks and Machine Learning (CNML)
Pages
853–856
Publication date
30 January 2026
Language
EN