Efficient Vehicular Collaborative Perception Based on Saptial-Temporal Feature Compression
IEEE Transactions on Vehicular Technology · 2024-11 · 73 (11) · 16125-16133
DOI: 10.1109/tvt.2024.3403263 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Vehicular Technology, 2024-11. Verfasst von Jingyu Zhang und 4 weiteren Autor:innen. Die Arbeit wurde bisher 17-mal zitiert (Crossref) und nennt 40 Literaturangaben. Gefördert von: Shanghai Key Research Laboratory of NSAI, the Innovation Platform for Academicians of Hainan Province, Haikou, the Specific Research Fund of the Innovation Platform for Academicians of Hainan Province.
Abstract (Originalsprache)
In collaborative perception, autonomous vehicles with limited perception capabilities can communicate with each other to achieve a more holographic and effective perception result. Real-world communication systems, however, are usually constrained by wireless communication resources or reliability and cannot handle the enormous real-time data transmissions to support the delay-sensitive collaborative perception. To resolve this issue, we propose an efficient vehicular Collaborative Perception method based on Spatial-temporal feature Compression (CPSC) that exploits the trade-off between perception performance and bandwidth consumption. It performs feature-level compression by focusing on critical regions of perceptual information in the spatial-temporal domain and adapts traffic according to network conditions. To the best of our knowledge, this paper presents the first work where temporal feature redundancy is considered for enhancing the efficiency of collaborative perception. To thoroughly evaluate CPSC, we conduct extensive experiments of collaborative 3D object detection tasks on the real-world dataset DAIR-V2X and the simulated dataset OPV2V. The results show that CPSC outperforms the SOTA collaborative perception methods by 2.91% for AP@0.7 in OPV2V dataset and 1.71% for AP@0.5 in DAIR-V2X dataset. Meanwhile, CPSC attains a communication volume reduction of more than 10 times while consistently outperforming the previous SOTA method.
Förderung
- Shanghai Key Research Laboratory of NSAI
- the Innovation Platform for Academicians of Hainan Province, Haikou
- the Specific Research Fund of the Innovation Platform for Academicians of Hainan Province
- National Natural Science Foundation of China
- Nanjing First Automobile Works Grant
Zitieren
APA
Zhang, J., Yang, K., Wang, H., Sun, P., & Song, L. (2024). Efficient Vehicular Collaborative Perception Based on Saptial-Temporal Feature Compression. IEEE Transactions on Vehicular Technology, 73(11), 16125-16133. https://doi.org/10.1109/tvt.2024.3403263
Vancouver
Zhang J, Yang K, Wang H, Sun P, Song L. Efficient Vehicular Collaborative Perception Based on Saptial-Temporal Feature Compression. IEEE Transactions on Vehicular Technology. 2024;73(11):16125-16133. doi:10.1109/tvt.2024.3403263
BibTeX
@article{Zhang2024,
author = {Zhang, Jingyu and Yang, Kun and Wang, Hanqi and Sun, Peng and Song, Liang},
title = {Efficient Vehicular Collaborative Perception Based on Saptial-Temporal Feature Compression},
journal = {IEEE Transactions on Vehicular Technology},
year = {2024},
volume = {73},
number = {11},
pages = {16125--16133},
doi = {10.1109/tvt.2024.3403263}
}
EUVIMED