EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Radiance Field Delta Video Compression in Edge-Enabled Vehicular Metaverse

Matus Dopiriak, Eugen Slapak, Juraj Gazda, Devendra Singh Gurjar, Mohammad Abdullah Al Faruque, Marco Levorato

IEEE Open Journal of Intelligent Transportation Systems · 2026 · 7 · 1977-1993

DOI: 10.1109/ojits.2026.3714093 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Open Journal of Intelligent Transportation Systems, 2026. Verfasst von Matus Dopiriak und 5 weiteren Autor:innen. Die Arbeit nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science Foundation, Ministry of Education, Science, Research, and Sport of the Slovak Republic, and Slovak Academy of Sciences, Agentúra na Podporu Výskumu a Vývoja.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Connected and autonomous vehicles (CAVs) offload computationally intensive tasks to multi-access edge computing (MEC) servers via vehicle-to-infrastructure (V2I) communication, enabling applications within the vehicular metaverse, which transforms physical environment into the digital space enabling advanced analysis or predictive modeling. A core challenge is physical-to-virtual (P2V) synchronization through digital twins (DTs), reliant on MEC networks and ultra-reliable low-latency communication (URLLC). To address this, we introduce radiance field (RF) delta video compression (RFDVC), which uses RF-encoder and RF-decoder architecture using distributed RFs as DTs storing photorealistic 3D urban scenes in compressed form. This method extracts differences between CAV-frame capturing actual traffic and RF-frame capturing empty scene from the same camera pose in batches encoded and transmitted over the MEC network. Experiments show data savings up to 71% against H.264 codec and 44% against H.265 codec under different conditions as lighting changes, and rain. RFDVC also demonstrates resilience to transmission errors, achieving up to +0.29 structural similarity index measure (SSIM) improvement at block error rate (BLER) $\mathcal{D}$ 0.35 in non-rainy conditions and $\mathcal{C}$ 0.25 at BLER $\mathcal{D}$ 0.2 in rainy conditions compared with standard video coding (VC) baselines. The latency analysis identifies three operating modes: full FastSAM reaches approximately 18 FPS, adaptive FastSAM reaches approximately 31 FPS, and no-FastSAM exceeds 45 FPS. Therefore, 30 FPS operation is achieved only in the adaptive and no-FastSAM configurations, whereas full FastSAM provides stronger class-agnostic delta segmentation at higher computational cost. The semantic IoU/Dice evaluation further indicates that the main foreground content is preserved after encoding and decoding. However, this analysis should be interpreted as a proxy for semantic consistency rather than as detector- or tracker-level downstream validation.

Förderung

  • National Science Foundation
  • Ministry of Education, Science, Research, and Sport of the Slovak Republic, and Slovak Academy of Sciences
  • Agentúra na Podporu Výskumu a Vývoja
  • EU NextGenerationEU through the Recovery and Resilience Plan for Slovakia

Zitieren

APA

Dopiriak, M., Slapak, E., Gazda, J., Gurjar, D. S., Faruque, M. A. A., & Levorato, M. (2026). Radiance Field Delta Video Compression in Edge-Enabled Vehicular Metaverse. IEEE Open Journal of Intelligent Transportation Systems, 7, 1977-1993. https://doi.org/10.1109/ojits.2026.3714093

Vancouver

Dopiriak M, Slapak E, Gazda J, Gurjar DS, Faruque MAA, Levorato M. Radiance Field Delta Video Compression in Edge-Enabled Vehicular Metaverse. IEEE Open Journal of Intelligent Transportation Systems. 2026;7:1977-1993. doi:10.1109/ojits.2026.3714093

BibTeX
@article{Dopiriak2026,
  author = {Dopiriak, Matus and Slapak, Eugen and Gazda, Juraj and Gurjar, Devendra Singh and Faruque, Mohammad Abdullah Al and Levorato, Marco},
  title = {Radiance Field Delta Video Compression in Edge-Enabled Vehicular Metaverse},
  journal = {IEEE Open Journal of Intelligent Transportation Systems},
  year = {2026},
  volume = {7},
  pages = {1977--1993},
  doi = {10.1109/ojits.2026.3714093}
}
Kostenlos starten →