Sat2Scene: 3D Urban Scene Generation from Satellite Images with Diffusion

TL;DR

This paper introduces Sat2Scene, a novel 3D scene generation method from satellite images using diffusion models and neural rendering, enabling realistic multi-view urban scene synthesis.

Contribution

It presents a new architecture combining diffusion models with neural rendering for direct 3D scene generation from satellite imagery, addressing view change and scale challenges.

Findings

Generates photo-realistic street-view sequences.

Produces consistent multi-view urban scenes.

Effective on city-scale datasets.

Abstract

Directly generating scenes from satellite imagery offers exciting possibilities for integration into applications like games and map services. However, challenges arise from significant view changes and scene scale. Previous efforts mainly focused on image or video generation, lacking exploration into the adaptability of scene generation for arbitrary views. Existing 3D generation works either operate at the object level or are difficult to utilize the geometry obtained from satellite imagery. To overcome these limitations, we propose a novel architecture for direct 3D scene generation by introducing diffusion models into 3D sparse representations and combining them with neural rendering techniques. Specifically, our approach generates texture colors at the point level for a given geometry using a 3D diffusion model first, which is then transformed into a scene representation in a feed-forward manner. The representation can be utilized to render arbitrary views which would excel in both single-frame quality and inter-frame consistency. Experiments in two city-scale datasets show that our model demonstrates proficiency in generating photo-realistic street-view image sequences and cross-view urban scenes from satellite imagery.