Geometry-guided Cross-view Diffusion for One-to-many Cross-view Image Synthesis

TL;DR

This paper introduces a geometry-guided diffusion model for one-to-many cross-view image synthesis, effectively handling the inherent uncertainty and geometric ambiguities between satellite and ground images.

Contribution

It proposes a novel geometry-guided cross-view condition strategy within diffusion models to improve diversity and accuracy in satellite-to-ground and ground-to-satellite image synthesis.

Findings

Outperforms baseline and state-of-the-art methods in quality and diversity

Achieves higher fidelity in synthesized images

Demonstrates effectiveness on three benchmark datasets

Abstract

This paper presents a novel approach for cross-view synthesis aimed at generating plausible ground-level images from corresponding satellite imagery or vice versa. We refer to these tasks as satellite-to-ground (Sat2Grd) and ground-to-satellite (Grd2Sat) synthesis, respectively. Unlike previous works that typically focus on one-to-one generation, producing a single output image from a single input image, our approach acknowledges the inherent one-to-many nature of the problem. This recognition stems from the challenges posed by differences in illumination, weather conditions, and occlusions between the two views. To effectively model this uncertainty, we leverage recent advancements in diffusion models. Specifically, we exploit random Gaussian noise to represent the diverse possibilities learnt from the target view data. We introduce a Geometry-guided Cross-view Condition (GCC) strategy to establish explicit geometric correspondences between satellite and street-view features. This enables us to resolve the geometry ambiguity introduced by camera pose between image pairs, boosting the performance of cross-view image synthesis. Through extensive quantitative and qualitative analyses on three benchmark cross-view datasets, we demonstrate the superiority of our proposed geometry-guided cross-view condition over baseline methods, including recent state-of-the-art approaches in cross-view image synthesis. Our method generates images of higher quality, fidelity, and diversity than other state-of-the-art approaches.