Can I read OmniRe: Omni Urban Scene Reconstruction on EtoBox?

OmniRe: Omni Urban Scene Reconstruction by Chen, Ziyu; Yang, Jiawei; Huang, Jiahui; de Lutio, Riccardo; Esturo, Janick Martinez; Ivanovic, Boris; Litany, Or; Gojcic, Zan; Fidler, Sanja; Pavone, Marco; Song, Li; Wang, Yue is a scholarly article available to read on EtoBox.

What is OmniRe: Omni Urban Scene Reconstruction about?

We introduce OmniRe, a holistic approach for efficiently reconstructing high-fidelity dynamic urban scenes from on-device logs. Recent methods for modeling driving sequences using neural radiance fields or Gaussian Splatting have demonstrated the potential of reconstructing challenging dynamic scenes, but often overlook pedestrians and other non-vehicle dynamic actors, hindering a complete pipeline for dynamic urban scene reconstruction. To that end, we propose a comprehensive 3DGS framework for driving scenes, named OmniRe, that allows for accurate, full-length reconstruction of diverse dynamic objects in a driving log. OmniRe builds dynamic neural scene graphs based on Gaussian representations and constructs multiple local canonical spaces that model various dynamic actors, including vehicles, pedestrians, and cyclists, among many others. This capability is unmatched by existing methods. OmniRe allows us to holistically reconstruct different objects present in the scene, subsequently enabling the simulation of reconstructed scenarios with all actors participating in real-time (~60Hz). Extensive evaluations on the Waymo dataset show that our approach outperforms prior state-of-the

Author

Chen, Ziyu; Yang, Jiawei; Huang, Jiahui; de Lutio, Riccardo; Esturo, Janick Martinez; Ivanovic, Boris; Litany, Or; Gojcic, Zan; Fidler, Sanja; Pavone, Marco; Song, Li; Wang, Yue

Published

2024

Language

EN