Metraipolis › archive › dc_d1e6ec2e4745

PRG-Fusion: Orchestrating Generative Priors with Reconstruction Evidence for Driving View Synthesis — arXiv:2609.06948

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.06948
License (as archived)
CC0-1.0
Archived
2026-10-16T15:09:28.478Z
SHA-256 of the archived text
ad09c3c84c4dcddbeef570243184593b2e8b68b9a5543ac8d6318b0a1b1fcebb
Stable URL of this page
https://metraipolis.com/documento/dc_d1e6ec2e4745

How to cite

PRG-Fusion: Orchestrating Generative Priors with Reconstruction Evidence for Driving View Synthesis — arXiv:2609.06948 — arXiv (en), https://arxiv.org/abs/2609.06948 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d1e6ec2e4745 (Metraipolis archive dc_d1e6ec2e4745, sha256 ad09c3c84c4dcddbeef570243184593b2e8b68b9a5543ac8d6318b0a1b1fcebb)
BibTeX
@misc{dcd1e6ec2e4745,
  title = {PRG-Fusion: Orchestrating Generative Priors with Reconstruction Evidence for Driving View Synthesis — arXiv:2609.06948},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.06948},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d1e6ec2e4745; sha256 ad09c3c84c4dcddbeef570243184593b2e8b68b9a5543ac8d6318b0a1b1fcebb},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.06948. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Synthesizing photorealistic driving videos along specified trajectories is essential for scalable closed-loop simulation. Reconstruction-based methods leverage neural rendering to synthesize geometrically consistent views, but often exhibit diverse artifacts and missing content when the viewpoint deviates from the training trajectory. In contrast, generative models can synthesize realistic views along arbitrary trajectories from vehicle sensor data, yet often struggle to maintain temporal and geometric consistency across frames. To combine the strengths of both, we propose PRG-Fusion, a framework for driving view synthesis that uses reconstruction evidence to orchestrate generative priors across regions. Specifically, we extract region-wise degradation evidence from reconstructed driving scenes and convert it into Preserve, Repair, and Generate (PRG) labels. At inference, these labels serve as a unified routing policy for region-aware spatiotemporal synthesis, orchestrating 3DGS appearance preservation, LiDAR-guided structural correction, and video-prior-driven content completion across Preserve, Repair, and Generate regions, respectively. We then follow a two-stage training paradigm, first establish geometric control from sparse LiDAR projections and subsequently learning appearance control from dense 3DGS renderings. Extensive experiments on Waymo demonstrate that PRG-Fusion achieves state-of-the-art overall performance in novel trajectory video synthesis, with superior visual quality and geometric fidelity while maintaining competitive view consistency under large trajectory shifts.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.06948 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T14:49:57.150Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).