Salvato in:
Dettagli Bibliografici
Autori principali: Chen, Zeming, Zhao, Hang
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2507.00707
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908429373669376
author Chen, Zeming
Zhao, Hang
author_facet Chen, Zeming
Zhao, Hang
contents Multi-view image generation in autonomous driving demands consistent 3D scene understanding across camera views. Most existing methods treat this problem as a 2D image set generation task, lacking explicit 3D modeling. However, we argue that a structured representation is crucial for scene generation, especially for autonomous driving applications. This paper proposes BEV-VAE for consistent and controllable view synthesis. BEV-VAE first trains a multi-view image variational autoencoder for a compact and unified BEV latent space and then generates the scene with a latent diffusion transformer. BEV-VAE supports arbitrary view generation given camera configurations, and optionally 3D layouts. Experiments on nuScenes and Argoverse 2 (AV2) show strong performance in both 3D consistent reconstruction and generation. The code is available at: https://github.com/Czm369/bev-vae.
format Preprint
id arxiv_https___arxiv_org_abs_2507_00707
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle BEV-VAE: Multi-view Image Generation with Spatial Consistency for Autonomous Driving
Chen, Zeming
Zhao, Hang
Computer Vision and Pattern Recognition
Multi-view image generation in autonomous driving demands consistent 3D scene understanding across camera views. Most existing methods treat this problem as a 2D image set generation task, lacking explicit 3D modeling. However, we argue that a structured representation is crucial for scene generation, especially for autonomous driving applications. This paper proposes BEV-VAE for consistent and controllable view synthesis. BEV-VAE first trains a multi-view image variational autoencoder for a compact and unified BEV latent space and then generates the scene with a latent diffusion transformer. BEV-VAE supports arbitrary view generation given camera configurations, and optionally 3D layouts. Experiments on nuScenes and Argoverse 2 (AV2) show strong performance in both 3D consistent reconstruction and generation. The code is available at: https://github.com/Czm369/bev-vae.
title BEV-VAE: Multi-view Image Generation with Spatial Consistency for Autonomous Driving
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2507.00707