Tian, D., Lin, C., Zhou, J., Duan, X., Cao, Y., Zhao, D. and Cao, D. (2022) SA-YOLOv3: an efficient and accurate object detector using self-attention mechanism for autonomous driving. IEEE Transactions on Intelligent Transportation Systems, 23(5), pp. 4099-4110. (doi: 10.1109/TITS.2020.3041278)
Text
302044.pdf - Accepted Version 15MB |
Abstract
Object detection is becoming increasingly significant for autonomous-driving system. However, poor accuracy or low inference performance limits current object detectors in applying to autonomous driving. In this work, a fast and accurate object detector termed as SA-YOLOv3, is proposed by introducing dilated convolution and self-attention module (SAM) into the architecture of YOLOv3. Furthermore, loss function based on GIoU and focal loss is reconstructed to further optimize detection performance. With an input size of 512×512 , our proposed SA-YOLOv3 improves YOLOv3 by 2.58 mAP and 2.63 mAP on KITTI and BDD100K benchmarks, with real-time inference (more than 40 FPS). When compared with other state-of-the-art detectors, it reports better trade-off in terms of detection accuracy and speed, indicating the suitability for autonomous-driving application. To our best knowledge, it is the first method that incorporates YOLOv3 with attention mechanism, and we expect this work would guide for autonomous-driving research in the future.
Item Type: | Articles |
---|---|
Status: | Published |
Refereed: | Yes |
Glasgow Author(s) Enlighten ID: | Zhao, Dr Dezong |
Authors: | Tian, D., Lin, C., Zhou, J., Duan, X., Cao, Y., Zhao, D., and Cao, D. |
College/School: | College of Science and Engineering > School of Engineering > Autonomous Systems and Connectivity |
Journal Name: | IEEE Transactions on Intelligent Transportation Systems |
Publisher: | IEEE |
ISSN: | 1524-9050 |
ISSN (Online): | 1558-0016 |
Published Online: | 17 December 2020 |
Copyright Holders: | Copyright © 2020 IEEE |
First Published: | First published in IEEE Transactions on Intelligent Transportation Systems 23(5):4099-4110 |
Publisher Policy: | Reproduced in accordance with the publisher copyright policy |
University Staff: Request a correction | Enlighten Editors: Update this record