SeasonStereo: Robust Dense Stereo Matching for Multi-Date Satellite Imagery via Generative AI

2026-07-29

AI Summary

This work addresses the challenge of dense stereo matching in multi-temporal satellite imagery, which is severely complicated by seasonal and illumination variations. The authors propose a novel approach that eliminates the need for near-simultaneous image pairs or ground-truth alignment labels. By training on synthetic image pairs with controllable seasonal changes and incorporating zero-shot geometric priors from foundation models, the method achieves high-accuracy disparity estimation using only unsupervised multi-temporal data. Experimental results demonstrate that the reconstruction accuracy rivals that of LiDAR-supervised models, while producing sharper geometric details. This framework enables high-quality 3D reconstruction from large-scale, heterogeneous satellite imagery with substantially reduced annotation costs.

Abstract

Accurate 3D reconstruction from satellite imagery typically relies on near-simultaneous stereo pairs, limiting its applicability to diachronic settings where multi-date images exhibit varying seasonal and illumination conditions. Training dense stereo matching models robust to appearance changes is a long-standing challenge, as aligned multi-date imagery and ground-truth geometry are costly to obtain at scale. We propose SeasonStereo, a scalable framework that addresses disparity estimation from diachronic satellite images by training on synthetic image pairs with controlled seasonal appearance variation, while leveraging zero-shot geometric priors from foundation models. SeasonStereo matches the accuracy of state-of-the-art LiDAR-supervised models, while producing sharper geometric details without requiring aligned real multi-date training products or LiDAR-derived labels. As a result, SeasonStereo offers a practical path toward large-scale 3D reconstruction from heterogeneous satellite images with reduced supervision cost.

Problem

• dense stereo matching
• multi-date satellite imagery
• seasonal variation
• 3D reconstruction
• appearance change

Innovation

• SeasonStereo
• dense stereo matching
• multi-date satellite imagery
• generative AI
• zero-shot geometric priors