Omnire : High-fidelity dynamic urban scene reconstruction technology

Omnire

AI image generation AI 3D tools #3D Reconstruction #Dynamic Scenes #Autonomous Driving #Virtual Reality Standard Picks Open Source

Overview :

OmniRe is a comprehensive method for efficiently reconstructing high-fidelity dynamic urban scenes from device logs. This technology achieves a complete reconstruction of different objects in the scene by constructing a dynamic neural scene graph based on Gaussian representations and building multiple local canonical spaces to simulate various dynamic actors, including vehicles, pedestrians, and cyclists. OmniRe enables comprehensive reconstruction of different objects present in a scene, allowing for real-time simulation of reconstructed scenes involving all participants. Extensive evaluations on the Waymo dataset show that OmniRe significantly outperforms previous state-of-the-art methods both quantitatively and qualitatively.

Target Users :

OmniRe is designed for urban planners, traffic engineers, autonomous vehicle technology researchers, and virtual reality content creators. It provides high-fidelity urban scene reconstruction, which is crucial for traffic simulation, urban planning, and testing autonomous driving systems.

Total Visits： 767

Website Views ： 51.9K

Use Cases

Simulate urban traffic flow to optimize traffic management strategies.

Simulate a complex urban environment during autonomous vehicle testing.

Create realistic urban backgrounds for virtual reality games and applications.

Features

Use neural radiance fields or Gaussian splatting to reconstruct dynamic scenes

Ignore pedestrians and other non-vehicle dynamic actors

Construct dynamic neural scene graphs based on Gaussian representations

Simulate rigid body transformations for vehicle movements

Fit near-distance walking pedestrians using a template-based SMPL model