UrbanIR: Large-Scale Urban Scene Inverse Rendering from a Single Video

TL;DR

UrbanIR is a novel inverse rendering framework that accurately infers scene geometry, materials, and lighting from a single wide-baseline video, enabling realistic free-viewpoint rendering and scene editing in urban environments.

Contribution

It introduces new loss functions and techniques for improved inverse graphics inference from challenging urban scene videos, outperforming existing methods in accuracy and realism.

Findings

Accurate shape, albedo, and illumination estimation from single videos.

Enhanced shadow volume and lighting control capabilities.

Superior rendering quality compared to prior state-of-the-art methods.

Abstract

We present UrbanIR (Urban Scene Inverse Rendering), a new inverse graphics model that enables realistic, free-viewpoint renderings of scenes under various lighting conditions with a single video. It accurately infers shape, albedo, visibility, and sun and sky illumination from wide-baseline videos, such as those from car-mounted cameras, differing from NeRF's dense view settings. In this context, standard methods often yield subpar geometry and material estimates, such as inaccurate roof representations and numerous 'floaters'. UrbanIR addresses these issues with novel losses that reduce errors in inverse graphics inference and rendering artifacts. Its techniques allow for precise shadow volume estimation in the original scene. The model's outputs support controllable editing, enabling photorealistic free-viewpoint renderings of night simulations, relit scenes, and inserted objects, marking a significant improvement over existing state-of-the-art methods.