EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Communication Efficient Cooperative Perception via Codebook-Free Vector Quantization

Junjie Wang, Zeyu Gao, Tomas Nordström

IEEE Access · 2026 · 14 · 42353-42365

DOI: 10.1109/access.2026.3674083 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Junjie Wang und 2 weiteren Autor:innen. Die Arbeit nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Part of the computations were enabled by resources provided by the National Academic Infrastructure for Supercomputing in Sweden (NAISS), partially, Vetenskapsrådet.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Recent advances in cooperative perception have demonstrated significant performance improvements over single-agent perception. In practice, cooperative perception methods often exchange intermediate multi-channel BEV feature maps at a high frame rate, which can impose a substantial burden on available V2X bandwidth and hinder the practical deployment of such systems. To mitigate bandwidth limitations, we introduce Codebook-Free Quantization for Cooperative Perception (CFQ4CP), a lightweight feature-compression framework. Specifically, a learned neural network, combined with a simple rounding operation, maps continuous latent features to compact integer codes. The discrete representations, instead of high-dimensional feature maps, are then transmitted through V2X communication. Subsequently, these features are reconstructed from the integer codes via a lightweight neural network compression decoder. Unlike existing compression methods that rely on fixed codebooks and nearest-neighbor lookups, our approach generates highly compact integer representations that eliminate the need for codebook alignment between transmitting and receiving agents. We evaluated the proposed framework on two real-world cooperative perception datasets, V2V4Real and DAIR-V2X. Experimental results demonstrate that CFQ4CP achieves perception accuracy competitive with full-precision models, even when each feature vector in the bird’s eye view map is quantized to extremely low bitwidths (e.g., 1 or 2 bits), results in reducing the communication costs by thousands of times.

Förderung

  • Part of the computations were enabled by resources provided by the National Academic Infrastructure for Supercomputing in Sweden (NAISS), partially
  • Vetenskapsrådet

Zitieren

APA

Wang, J., Gao, Z., & Nordström, T. (2026). Communication Efficient Cooperative Perception via Codebook-Free Vector Quantization. IEEE Access, 14, 42353-42365. https://doi.org/10.1109/access.2026.3674083

Vancouver

Wang J, Gao Z, Nordström T. Communication Efficient Cooperative Perception via Codebook-Free Vector Quantization. IEEE Access. 2026;14:42353-42365. doi:10.1109/access.2026.3674083

BibTeX
@article{Wang2026,
  author = {Wang, Junjie and Gao, Zeyu and Nordström, Tomas},
  title = {Communication Efficient Cooperative Perception via Codebook-Free Vector Quantization},
  journal = {IEEE Access},
  year = {2026},
  volume = {14},
  pages = {42353--42365},
  doi = {10.1109/access.2026.3674083}
}
Kostenlos starten →